[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-794":3,"related-794":76},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":38,"search_phrases":70,"slug":73,"view_count":74,"doi":8,"paper":8,"created_at":75},794,"方向对齐惩罚(DAP):改进YOLOv8边界框回归的轻量化定位损失(原标题:Directional Alignment Penalty: A Lightweight Localization Loss for Improved Bounding Box Regression in YOLOv8)","https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286","精准农业视觉流水线下游任务(形态特征提取、生长监测、数字孪生同步)直接依赖检测框的几何质量,因此精确定位(而非单纯正确分类)成为关键性能因素。本研究提出方向对齐惩罚(DAP)作为辅助定位正则化项,在不修改检测器架构的情况下将独立加权的归一化中心位移项引入边界框回归目标。DAP-YOLOv8使mAP@0.5:0.95从51.69%提升至53.76%、mAP@0.5从80.03%提升至80.36%,在真实世界智能温室采集的牡蛎蘑菇数据集上重复种子实验进一步表明细粒度定位的提升在不同随机初始化下一致。",null,"## 1. Introduction\n\nSmart agriculture applications rely on computer-vision systems as the backbone for continuous visual monitoring, enabling tasks such as yield estimation, disease detection, autonomous harvesting, and environmental control. Object detection is rarely the final goal of these pipelines: the detected bounding boxes are an intermediate representation from which morphological descriptors (object size, count, maturity, and spatial distribution) are automatically derived and fused with other data streams. Thus, the quality of these descriptors is limited by the geometric accuracy of the detected boxes, and accurate localization (rather than just correct classification) becomes the key performance factor.\n\nThe YOLO family of single-stage detectors [[1](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B1-applsci-16-08286)] is widely used in this area because it predicts object locations and class probabilities in a single forward pass, providing real-time inference with competitive accuracy. In particular, YOLOv8 [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] combines a decoupled detection head, a multi-scale feature aggregation neck and a hierarchical backbone, and has proven effective for continuous agricultural monitoring. The localization component of such detectors is driven by a bounding-box regression loss, and the design of this loss has a strong influence on both the convergence speed and the final localization quality.\n\nMost modern regression losses are formulated based on the intersection-over-union (IoU) between the predicted boxes and ground-truth boxes. To improve gradient behavior, IoU-based variants such as GIoU [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], DIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286)] and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] add geometric constraints based on an enclosing area, a center-distance term or an aspect-ratio penalty. However, in DIoU and CIoU, the normalized center-distance term is integrated into a single objective function with an implicit weight of one (or, for the aspect-ratio term of CIoU, with an adaptive trade-off coefficient), so it cannot be tuned, scheduled or monitored independently of the overlap and aspect-ratio terms. This limitation is most pronounced in agricultural scenes, where objects usually have irregular contours, partial occlusion and dense spatial clustering, as exemplified by the detection of oyster mushroom clusters in a cultivation chamber. To this end, we propose a Directional Alignment Penalty (DAP), an additional lightweight regularization term that explicitly encourages the predicted boxes to move toward their corresponding ground-truth boxes. DAP is measured as the Euclidean distance between the predicted and ground-truth box centers, normalized by the diagonal of the smallest box encapsulating both. We show in [Section 3.4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec3dot4-applsci-16-08286) that this quantity is, up to a square root, exactly the center-distance factor already used inside DIoU and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)]. The contribution of DAP is to decouple that factor from the composite loss and re-express it as a separately weighted additive term whose effect on optimization can be tuned and logged independently of the base regression loss the detector already uses.\n\nThis work makes three contributions. We reformulate the normalized center-distance factor used in DIoU\u002FCIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] as an independently weighted additive regularization term that can be incorporated into existing localization losses without modifying the detector architecture. Second, we demonstrate a practical integration of the penalty into the YOLOv8 training and validation pipeline, including the separate logging of the penalty term for interpretability. Third, we evaluate the method on a dedicated oyster mushroom (Pleurotus ostreatus) dataset collected from a real-world smart greenhouse and demonstrate that stacking DAP on top of the existing CIoU-based box loss of YOLOv8 results in an additional measurable improvement in fine-grained localization, most notably a 2.07 percentage-point improvement in mAP@0.5:0.95, while preserving precision, recall, and the computational efficiency of the base detector.\n\nThe remainder of the paper is organized as follows. [Section 2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec2-applsci-16-08286) reviews related work on single-stage detection and bounding-box regression losses. In [Section 3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec3-applsci-16-08286), we describe the dataset, the YOLOv8 detector, the proposed DAP formulation, its formal relation to DIoU and CIoU, its integration into the detector, and the training protocol. The experimental results are presented in [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286). The implications and limitations of the results are discussed in [Section 5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec5-applsci-16-08286). [Section 6](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec6-applsci-16-08286) concludes the paper.\n\n## 2. Related Work\n\n### 2.1. Deep Learning-Based Object Detection\n\nDeep learning has revolutionized object detection by replacing hand-engineered feature extraction with end-to-end trainable convolutional neural networks (CNNs). Modern object detectors are commonly classified into two-stage and one-stage architectures. Two-stage detectors such as Faster R-CNN first generate region proposals and then classify and refine them, resulting in high localization accuracy but at the cost of increased computational complexity [[1](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B1-applsci-16-08286)]. In contrast, one-stage detectors perform object localization and classification in a single step, enabling substantially faster inference and making them suitable for real-time applications [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)].\n\nAmong one-stage detectors, the YOLO (You Only Look Once) family of detectors has become one of the most influential object detection frameworks due to its excellent speed-accuracy trade-off. The original YOLO architecture [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] and its subsequent versions have continuously improved feature extraction, label assignment, multi-scale representation and localization accuracy. The most recent version, YOLOv8, is now one of the most popular object detectors in real-time computer vision applications [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], thanks to its anchor-free detection approach, decoupled detection heads, improved feature fusion, and optimized training processes. However, despite these architectural improvements, the overall detection performance of YOLOv8 continues to depend strongly on the effectiveness of its bounding-box regression loss.\n\n### 2.2. IoU-Based Bounding Box Regression Losses\n\nBounding-box regression is critical to object detection because it directly determines localization accuracy. Early object detectors used coordinate-based regression losses (e.g., L1 loss and Smooth L1 loss), which considered each coordinate independently. These losses are computationally efficient but correlate weakly with the actual localization quality as measured by the Intersection over Union (IoU) metric. To overcome this limitation, Rezatofighi et al. proposed Generalized IoU (GIoU) [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], which extends the conventional IoU metric by accounting for the smallest enclosing box and can be optimized even when the predicted and ground-truth boxes do not overlap. Later, Zheng et al. proposed Distance IoU (DIoU) to speed up convergence and improve localization accuracy by adding the normalized Euclidean distance between the centers of bounding boxes [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286)]. The same authors further proposed Complete IoU (CIoU) [[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] by incorporating aspect-ratio consistency, resulting in faster convergence and better localization performance than previous IoU-based losses.\n\nBased on the success of DIoU and CIoU, many variants have been proposed to further improve bounding-box regression. Alpha-IoU [[6](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B6-applsci-16-08286)] introduces a power transformation of the IoU term to enable tunable optimization of different localization accuracies and improve robustness against noisy annotations and small datasets. MPDIoU [[7](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B7-applsci-16-08286)] replaces the classical center-distance constraint with the minimum point distances between the corresponding corners of the predicted and ground-truth boxes, which is simpler to compute and also considers the overlap, center displacement, and width-height difference [[7](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B7-applsci-16-08286)]. Among these methods, DIoU and CIoU already include a normalized center-distance term, but this term is merged into a single composite objective with an implicit, fixed weight, so its contribution to optimization is not tunable, schedulable, or monitorable independently of the overlap and aspect-ratio components. Alpha-IoU and MPDIoU jointly reweight or restructure the geometric factors, but the center-alignment contribution is still entangled with the rest of the loss. This limitation is particularly relevant for small, densely distributed, or partially occluded objects, for which the relative weight assigned to center alignment during optimization can be as important as the geometric factor itself.\n\nMotivated by this observation, we revisit the normalized center-distance term already present in DIoU and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] and reformulate it as the Directional Alignment Penalty (DAP): an externally added, independently weighted, unsquared regularization component that penalizes the normalized displacement between predicted and ground-truth bounding-box centers. The contribution of DAP is not to introduce a new geometric localization signal but to decouple this factor from the composite DIoU\u002FCIoU formulation. This decoupling allows its weight to be controlled, scheduled, and logged independently of the base regression loss while maintaining full compatibility with existing IoU-based regression losses.\n\n### 2.3. Object Detection in Precision Agriculture\n\nPrecision agriculture has become one of the most promising application domains of artificial intelligence by enabling data-driven crop monitoring, disease diagnosis, yield estimation, and autonomous harvesting. The rapid development of computer vision and deep learning has significantly improved the ability of agricultural systems to detect, classify, and monitor agricultural products under real-world environmental conditions. Compared with traditional machine vision methods based on handcrafted features, deep learning-based object detectors provide greater robustness to variations in illumination, object scale, occlusion, and background complexity, making them highly suitable for agricultural applications [[8](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B8-applsci-16-08286),[9](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B9-applsci-16-08286)].\n\nRecent studies have demonstrated the successful application of deep learning-based object detection across numerous agricultural tasks, including fruit counting, crop growth monitoring, weed identification, disease detection, livestock monitoring, and greenhouse automation [[10](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B10-applsci-16-08286),[11](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B11-applsci-16-08286)]. Among the available detection frameworks, the YOLO family has become particularly attractive because of its favorable trade-off between detection accuracy and computational efficiency. Consequently, various YOLO-based architectures have been adopted for detecting apples, tomatoes, strawberries, cucumbers, peppers, grapes, citrus fruits, wheat heads, rice panicles, and other agricultural products under challenging field conditions [[12](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B12-applsci-16-08286),[13](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B13-applsci-16-08286),[14](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B14-applsci-16-08286)]. These applications require detectors capable of maintaining high localization accuracy despite illumination changes, cluttered backgrounds, overlapping objects, and varying viewing angles.\n\nOne of the most challenging agricultural detection problems is mushroom monitoring in greenhouse environments. Unlike many fruits that are relatively well separated, oyster mushrooms typically grow in dense clusters, exhibit irregular shapes, and frequently overlap at different growth stages. In addition, variations in illumination, shadows, and moisture, together with occlusion by neighboring mushrooms, further complicate accurate localization. These characteristics often reduce the effectiveness of conventional object detectors because even small localization errors may lead to inaccurate size estimation, incorrect maturity assessment, or missed detections during automated harvesting [[15](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B15-applsci-16-08286)].\n\nSeveral recent studies have explored the application of deep learning to mushroom detection and yield estimation. YOLO-based detectors have demonstrated promising performance in identifying mushroom caps, estimating maturity stages, and supporting automated harvesting systems. Researchers have also investigated improvements in feature extraction, attention mechanisms, data augmentation, and image preprocessing to increase detection accuracy under greenhouse conditions [[16](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B16-applsci-16-08286),[17](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B17-applsci-16-08286),[18](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B18-applsci-16-08286),[19](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B19-applsci-16-08286)]. Despite these advances, relatively little attention has been devoted to improving the bounding-box regression objective, even though localization quality directly influences downstream measurements such as mushroom diameter estimation, growth analysis, and digital twin-based monitoring systems. Bounding-box localization is particularly important in mushroom cultivation because multiple mushrooms frequently appear in close proximity and have partially overlapping boundaries. Under these conditions, conventional IoU-based regression losses may produce predicted bounding boxes with similar overlap scores despite differing spatial alignments with the ground truth. Such localization inconsistencies can propagate to subsequent processing stages, affecting object tracking, phenotypic measurements, and yield estimation. Therefore, improving the regression objective rather than modifying the detector architecture offers a lightweight yet effective strategy for enhancing localization accuracy while preserving real-time performance.\n\nMotivated by these difficulties, this work aims to enhance the localization capability of YOLOv8 by reinterpreting the normalized center-distance factor in DIoU\u002FCIoU as the Directional Alignment Penalty (DAP), an additional, independently weighted, unsquared regularization term rather than a novel geometric localization signal. Instead of adding network modules or increasing model complexity, DAP augments the regression loss with this decoupled and independently weighted term. Since the proposed method is only active during training and does not modify the inference pipeline, it maintains the computational efficiency of YOLOv8 while delivering more accurate bounding-box localization, as shown by the observed gains in fine-grained localization ([Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286)). This feature makes DAP particularly well-suited for precision agriculture applications where lightweight models deployable in real time on embedded platforms are highly desirable.\n\n## 3. Materials and Methods\n\n### 3.1. Dataset\n\nThe oyster mushroom (Pleurotus ostreatus) image dataset was specifically constructed in a real-world smart mushroom greenhouse equipped with an RGB camera system and IoT-based monitoring infrastructure [[15](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B15-applsci-16-08286)]. Unlike many of the existing mushroom datasets collected under laboratory conditions or with sparse annotations, this dataset was designed to capture mushrooms in production environments, under variations in illumination, growth stage, object density, and environmental conditions. Images were acquired from two consecutive cultivation cycles in a greenhouse in Mersin, Türkiye, between late December 2022 and late April 2023 using two Full-HD (1920 × 1080) IP cameras. Images were taken during the day and at night to capture varied growth stages and lighting conditions. Each image was time-stamped to match environmental sensor readings.\n\nDuring the two cultivation cycles described above, about 1200 raw greenhouse frames were collected. After frames with no visible mushrooms and images of poor visual quality were excluded, 555 representative full-frame images were retained for object-detection experiments. These images were then split into training and validation subsets containing 499 images and 56 images, respectively. Bounding-box annotations were generated for mushroom instances visible in these full-frame images using the open-source LabelMe [[20](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B20-applsci-16-08286)] tool. Each bounding box was then refined to tightly enclose the mushroom while including minimal background. This procedure improved localization quality during supervised training. One annotation in the training set was excluded from the released label set because its class ID (2) did not match the declared two-class (immature\u002Fmature) configuration; this record was removed before training, rather than remapped, because the correct class could not be determined retrospectively.\n\nA separate, substantially larger set of approximately 8282 maturity-labeled samples was also derived from the same underlying image collection by manually cropping individual mushrooms into separate images assigned to two maturity classes, immature (5124) and mature (3158), with a balanced distribution of day (~4000) and night (~3963) instances. These cropped samples support downstream maturity-related analyses (e.g., per-instance maturity classification) and should not be interpreted as the number of full-frame images used to train the YOLO detector reported in this paper; the YOLOv8 and DAP-YOLOv8 localization experiments in [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286) used only the 555-image, full-frame detection dataset (499 training\u002F56 validation) described above. Annotations for the full-frame detection dataset were exported in PASCAL VOC (XML), COCO (JSON), and YOLO Darknet (TXT) formats for compatibility with several detection frameworks.\n\n### 3.2. YOLOv8 Detector\n\nYOLOv8 [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] was adopted as the base detector because its single-pass design maintains high localization accuracy while greatly reducing computational cost, thereby meeting the real-time monitoring and digital-t","MDPI Applied Sciences 16(16):8286 2026-08-20","2026-08-20T00:00:00Z","论文",10,false,71,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":22},12,20,18,9,1,"针对农业场景提出改进YOLOv8定位损失的新方法，有具体实验支撑，但属于细分技术进展，影响范围有限。",[24],{"name":10,"url":6},[26,27,28,29,30],"精准农业","YOLOv8","边界框回归","方向对齐惩罚","智能温室",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture","B",[],{"status":39,"locked":14,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":47},[44,52,58,65],{"code":45,"score":46,"supported":47,"title":48,"abstract":49,"tags":51},"I1",7,true,[],[26,50],"数字孪生",[26],{"code":32,"score":53,"supported":47,"title":54,"abstract":55,"tags":57},4,[],[56],"温室",[30],{"code":59,"score":60,"supported":47,"title":61,"abstract":62,"tags":64},"B5",3,[],[63],"蘑菇",[],{"code":66,"score":21,"supported":14,"title":67,"abstract":68,"tags":69},"I2",[],[],[27],[71,72],"方向对齐惩罚 边界框回归 智能温室 精准农业","方向对齐惩罚 边界框回归","方向对齐惩罚边界框回归智能温室精准农业-794",2,"2026-08-21T00:05:57.684831Z",{"total":77,"page":21,"page_size":77,"items":78},6,[79,155,237,331,397,447],{"id":80,"title":81,"url":82,"summary":83,"summary_zh":84,"content":8,"source_name":85,"source_url":82,"published_at":86,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":87,"score_detail":88,"sources":92,"tags":94,"primary_topic":99,"secondary_topics":100,"classification":107,"search_phrases":127,"slug":130,"view_count":131,"doi":132,"paper":133,"created_at":154},5237,"Integrasi Framework Laravel dan Android pada Sistem IoT untuk Monitoring Parameter Fisika-Kimia di Greenhouse","https:\u002F\u002Fdoi.org\u002F10.36257\u002Fteknika.v20i3.12040","Pengembangan teknologi pertanian presisi (precision agriculture) pada ekosistem greenhouse menggunakan sistem hidroponik menghadapi tantangan teknis dalam menjaga stabilitas parameter lingkungan mikro secara simultan. Penelitian ini bertujuan untuk mengintegrasikan framework Laravel dan platform Android ke dalam sistem Internet of Things (IoT) untuk memantau parameter fisika-kimia lingkungan, meliputi suhu udara, kelembaban, intensitas cahaya, dan kadar karbon dioksida (CO2), pada budidaya pakcoy (Brassica rapa L.) di greenhouse hidroponik. Fokus utama penelitian ini terletak pada perancangan arsitektur backend RESTful API menggunakan Laravel untuk mengelola penyimpanan basis data riwayat jangka panjang serta menyinkronkan transmisi data dari multi-sensor (DHT22, LDR, dan sensor gas CO2) yang memiliki karakteristik sampling rate berbeda. Di sisi pengguna, aplikasi Android dan antarmuka web dikembangkan sebagai frontend yang mengonsumsi data melalui serialisasi format JSON menggunakan metode polling HTTP GET real-time, serta menyediakan tiga mode kendali aktuator, yaitu manual, otomatis berbasis ambang batas sensor, dan penjadwalan. Pengujian fungsionalitas dan kinerja sistem dilakukan melalui visualisasi antarmuka pada dasbor situs web neliti-sk.darmajaya.ac.id serta evaluasi transmisi data nirkabel. Infrastruktur backend berbasis Laravel terbukti stabil dalam menangani fluktuasi frekuensi data sensor yang kompleks. Berdasarkan evaluasi log data selama 24 jam pengujian kontinu pada rentang 14 Agustus 2026 (00.00–23.00 WIB), sistem berhasil merekam seluruh 24 titik data per jam (100%) tanpa satupun jam yang terlewat. Integrasi ini menghadirkan solusi pemantauan sekaligus pengendalian pertanian presisi yang andal, terintegrasi, dan memiliki skalabilitas tinggi bagi operator greenhouse, meskipun masih memerlukan mekanisme offline logging untuk mengantisipasi gangguan konektivitas atau pemadaman listrik di masa mendatang.","在温室生态系统中利用水培系统开发精准农业（precision agriculture）技术，面临着同时维持微环境参数稳定性的技术挑战。本研究旨在将Laravel框架和Android平台集成到物联网（IoT）系统中，以监测水培温室中小白菜（Brassica rapa L.）栽培环境的物理化学参数，包括空气温度、湿度、光照强度和二氧化碳（CO2）浓度。本研究的主要重点在于设计基于Laravel的RESTful API后端架构，以管理长期历史数据库的存储，并同步具有不同采样率特征的多传感器（DHT22、LDR和CO2气体传感器）的数据传输。在用户端，开发了Android应用和Web界面作为前端，通过HTTP GET实时轮询方法以JSON格式序列化消费数据，并提供三种执行器控制模式，即手动、基于传感器阈值的自动控制和定时调度。系统功能与性能测试通过neliti-sk.darmajaya.ac.id网站仪表板的界面可视化以及无线数据传输评估进行。基于Laravel的后端基础设施在处理复杂的传感器数据频率波动方面表现出稳定性。根据2026年8月14日（00:00–23:00 WIB）连续24小时测试期间的数据日志评估，系统成功记录了每小时全部24个数据点（100%），无一小时遗漏。该集成提供了一种可靠、集成且具有高可扩展性的精准农业监测与控制解决方案，适用于温室操作人员，但未来仍需离线日志记录机制以应对连接中断或停电情况。","TEKNIKA","2026-10-17T00:00:00Z",56,{"impact":89,"substance":19,"depth":90,"authority":77,"freshness":89,"relevant":21,"comment":91},8,16,"论文将Laravel与Android集成用于水培温室多参数监测，方法具体、数据连续24小时完整，但属单点应用研究，公共影响有限，适合作为技术案例而非每日精选头条。",[93],{"name":85,"url":82},[95,96,26,97,98],"智慧农业","物联网","温室大棚","水培蔬菜",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[101],{"code":102,"slug":103,"name":104,"path":105,"parent_code":106},"I5","agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot","I",{"status":39,"locked":14,"version":40,"method":41,"evidence":108},{"candidates":109,"agricultural_evidence":47},[110,116,122],{"code":102,"score":19,"supported":47,"title":111,"abstract":113,"tags":115},[112],"IoT",[96,114,112],"internet of things",[96],{"code":32,"score":117,"supported":47,"title":118,"abstract":120,"tags":121},14,[119],"greenhouse",[56,119],[],{"code":45,"score":89,"supported":47,"title":123,"abstract":124,"tags":126},[],[26,125],"precision agriculture",[95,26],[128,129],"Laravel Android 温室物联网","小白菜 水培 温室监测","LaravelAndroid温室物联网-5237",0,"10.36257\u002Fteknika.v20i3.12040",{"doi":132,"openalex_id":134,"authors":135,"venue":85,"cited_by_count":131,"oa_url":146,"card":147,"direction":151,"ingested_from":153},"W7222850582",[136,138,141,144],{"name":137,"orcid":8},"Ramadanzah Ramadanzah",{"name":139,"orcid":140},"Dodi Yudo Setyawan","https:\u002F\u002Forcid.org\u002F0000-0003-2205-7617",{"name":142,"orcid":143},"Nurfiana Nurfiana","https:\u002F\u002Forcid.org\u002F0000-0002-6149-0918",{"name":145,"orcid":8},"Refi Arioen","https:\u002F\u002Fjurnal.polsri.ac.id\u002Findex.php\u002Fteknika\u002Farticle\u002Fdownload\u002F12040\u002F4368",{"tldr":148,"method":149,"finding":150,"direction":151,"opportunity":152},"集成Laravel与Android的IoT系统，实时监测温室水培小白菜的温湿度、光照和CO2。","用Laravel构建RESTful API后端，Android与Web前端通过J","后端稳定处理异构采样率数据，24小时连续测试每小时数据完整记录率100%。","智慧农业 \u002F 农业物联网","可研究断网或断电下的离线缓存与数据补传机制，提升温室IoT系统鲁棒性。","openalex","2026-10-18T23:30:07.964809Z",{"id":156,"title":157,"url":158,"summary":159,"summary_zh":160,"content":8,"source_name":161,"source_url":158,"published_at":162,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":163,"score_detail":164,"sources":168,"tags":170,"primary_topic":174,"secondary_topics":175,"classification":176,"search_phrases":207,"slug":210,"view_count":21,"doi":211,"paper":212,"created_at":236},4984,"A Comparative Review of Soilless Precision Agriculture Techniques and Emerging Smart Farming Technologies","https:\u002F\u002Fdoi.org\u002F10.38094\u002Fjastt721251","The rapidly diminishing arable land, uncertain climatic conditions, and lack of ample natural resources are worsening the urgency for adopting sustainable agricultural systems. Soilless cultivation methods promote enhanced resource utilization and better productivity. These methods, like aeroponics, hydroponics, and aquaponics, are evolving as smart farming techniques, providing notable advantages over traditional farming methods, enhancing water conservation, resource utilization, and enabling cultivation in places with reduced area and less water. The concept of soilless cultivation provides multifold advantages over traditional cultivation methods by further making provision for cultivating in the closest proximity to urban farmers, like residential rooftops. A bibliometric analysis is also conducted to highlight the recent trends in smart precision farming technologies and the choice of substrate for each technique. In this article, various methods of soilless smart precision agriculture are identified and compared with traditional cultivation methods. Moreover, aspects like economic sustainability and recent research trends are analyzed to obtain insight into the three basic classes of soilless cultivation, namely aeroponics, aquaponics, and hydroponics. Furthermore, the study also focuses on briefly analyzing the economics of soilless techniques. The study concludes by outlining the major challenges and possible future enhancements in this domain that can lead to the evolution of smart, efficient, and resilient agricultural systems.","可耕地迅速减少、气候条件不稳定以及充足自然资源的匮乏，正日益加剧采用可持续农业体系的紧迫性。无土栽培方法促进了资源利用率的提高和生产力的提升。气雾栽培、水培和鱼菜共生等方法正演变为智慧农业技术，相较传统耕作方式具有显著优势，能够增强节水能力、提高资源利用率，并使得在面积有限、水资源匮乏的地区进行种植成为可能。无土栽培的概念相较传统栽培方法具有多重优势，还能进一步满足在城市农户附近（如住宅屋顶）进行种植的需求。本文还进行了文献计量分析，以揭示智慧精准农业技术的最新趋势以及每种技术对基质的选择。本文识别了多种无土智慧精准农业方法，并将其与传统栽培方法进行比较。此外，还分析了经济可持续性和近期研究趋势等方面，以深入了解无土栽培的三大基本类别，即气雾栽培、鱼菜共生和水培。进一步地，本研究还着重简要分析了无土栽培技术的经济性。研究最后概述了该领域的主要挑战及可能的未来改进方向，这些改进有望推动智慧、高效且具有韧性的农业体系的发展。","Journal of Applied Science and Technology Trends","2026-10-14T00:00:00Z",76,{"impact":19,"substance":18,"depth":165,"authority":166,"freshness":89,"relevant":21,"comment":167},17,13,"系统比较无土栽培三类技术与智慧农业趋势，含文献计量分析，对设施农业与都市农业有参考价值。",[169],{"name":161,"url":158},[95,26,171,172,173],"垂直农业","水培","无土栽培",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":177},{"candidates":178,"agricultural_evidence":47},[179,184,188,195,201],{"code":45,"score":180,"supported":47,"title":181,"abstract":182,"tags":183},19,[125],[95,26,125],[95,26],{"code":32,"score":77,"supported":47,"title":185,"abstract":186,"tags":187},[],[173],[171,173,172],{"code":189,"score":77,"supported":47,"title":190,"abstract":191,"tags":194},"I7",[],[192,193],"文献计量","bibliometric",[],{"code":196,"score":60,"supported":47,"title":197,"abstract":198,"tags":200},"H3",[],[199],"水资源",[],{"code":202,"score":60,"supported":47,"title":203,"abstract":204,"tags":206},"H8",[],[205],"可持续农业",[],[208,209],"无土栽培 智慧农业 技术对比","水培 气雾培 鱼菜共生 精准农业","无土栽培智慧农业技术对比-4984","10.38094\u002Fjastt721251",{"doi":211,"openalex_id":213,"authors":214,"venue":161,"cited_by_count":131,"oa_url":158,"card":231,"direction":151,"ingested_from":153},"W7221762528",[215,218,221,223,226,229],{"name":216,"orcid":217},"Monica Dutta","https:\u002F\u002Forcid.org\u002F0000-0002-6235-9794",{"name":219,"orcid":220},"Deepali Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-6771-2534",{"name":222,"orcid":8},"Walaa Jamal Alharthi",{"name":224,"orcid":225},"Mohammad Zubair Khan","https:\u002F\u002Forcid.org\u002F0000-0002-2409-7172",{"name":227,"orcid":228},"Adnan Nadeem","https:\u002F\u002Forcid.org\u002F0000-0001-9622-150X",{"name":230,"orcid":8},"Sapna Juneja",{"tldr":232,"method":233,"finding":234,"direction":151,"opportunity":235},"综述无土精准农业技术，比较气雾培、水培和鱼菜共生，并分析智能农业趋势。","文献综述与文献计量分析，比较无土栽培技术、基质选择及经济可持续性。","无土栽培在节水、资源利用和城市近郊种植上优于传统农业，但经济性和技术挑战仍存。","可研究无土栽培与物联网、AI决策模型的深度融合，优化基质选择和能耗经济性。","2026-10-15T23:30:08.336176Z",{"id":238,"title":239,"url":240,"summary":241,"summary_zh":242,"content":8,"source_name":243,"source_url":240,"published_at":244,"category":12,"cover_url":8,"hotness":13,"is_selected":47,"score":245,"score_detail":246,"sources":249,"tags":251,"primary_topic":255,"secondary_topics":256,"classification":262,"search_phrases":302,"slug":305,"view_count":21,"doi":306,"paper":307,"created_at":330},4371,"Intelligent greenhouse irrigation control via multilayer perceptron neural network climate prediction integrated with fuzzy logic control","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compeleceng.2026.111587","Conventional irrigation methods rely on indoor sensor feedback, which fails to anticipate rapid microclimate fluctuations and often leads to over-irrigation or water stress. Additionally, traditional controllers used in the watering process lack the ability to adjust water flow rate dynamically, resulting in inefficient water usage and energy waste. Intelligent Greenhouse Irrigation Management (IGIM) remains a critical challenge for achieving sustainable, high-yield greenhouse agriculture. Precise irrigation management is especially important for crops like tomatoes, where both timing and quantity directly influence productivity and energy consumption. To address these challenges, this paper proposes a hybrid intelligent system that integrates a Multi-Layer Perceptron Artificial Neural Network (MLP-ANN) with an Adaptive Fuzzy Logic Controller (AFLC). The MLP-ANN forecasts key indoor variables such as temperature, humidity, and soil moisture, achieving R² values of 0.999. The Root Mean Square Errors of 0.266 °C, 0.254%, and 0.808%, respectively. These predictions directly feed into the AFLC, which employs an adaptive rule base to dynamically adjust irrigation flow rates in real time (0–18 LPH), significantly enhancing water-use efficiency and reducing waste to save energy consumption in the watering process. The key novelty of this work is the direct integration of MLP-predicted microclimate variables into the antecedent of adaptive fuzzy rules, enabling the controller to anticipate and respond to climate variations before they affect crop conditions. Unlike conventional reactive or static rule-based systems, the proposed MLP-AFLC framework creates a closed-loop coupling between forecasting and actuation, allowing predictive, rather than threshold-based, irrigation decisions. By integrating predictive intelligence with adaptive control, the proposed MLP-AFLC framework provides a robust, autonomous, and sustainable solution for greenhouse management, advancing precision agriculture in highly dynamic environments.","传统灌溉方法依赖室内传感器反馈，无法预判微气候的快速波动，常导致过度灌溉或水分胁迫。此外，灌溉过程中使用的传统控制器缺乏动态调节流速的能力，造成水资源利用效率低下和能源浪费。智能温室灌溉管理（IGIM）仍是实现可持续、高产温室农业的关键挑战。对于番茄等作物而言，精准灌溉管理尤为重要，因为灌溉时机和水量直接影响产量和能耗。为解决这些挑战，本文提出了一种将多层感知器人工神经网络（MLP-ANN）与自适应模糊逻辑控制器（AFLC）相集成的混合智能系统。MLP-ANN预测温度、湿度和土壤湿度等关键室内变量，R²值达到0.999，均方根误差分别为0.266 °C、0.254%和0.808%。这些预测结果直接输入AFLC，后者采用自适应规则库实时动态调节灌溉流速（0–18 LPH），显著提高水资源利用效率并减少浪费，从而在灌溉过程中节约能耗。本工作的关键创新在于将MLP预测的微气候变量直接集成到自适应模糊规则的前件中，使控制器能够在气候变化影响作物状况之前进行预判和响应。与传统反应式或静态规则系统不同，所提出的MLP-AFLC框架在预测与执行之间建立了闭环耦合，实现了基于预测而非基于阈值的灌溉决策。通过将预测智能与自适应控制相结合，所提出的MLP-AFLC框架为温室管理提供了一种稳健、自主且可持续的解决方案，推动了高度动态环境下的精准农业发展。","Computers & Electrical Engineering","2026-10-06T00:00:00Z",81,{"impact":19,"substance":247,"depth":19,"authority":117,"freshness":20,"relevant":21,"comment":248},22,"提出MLP-ANN与自适应模糊控制融合的温室预测性灌溉框架，R²达0.999，方法新颖且数据可靠，对精准农业节水节能有实质参考价值。",[250],{"name":243,"url":240},[95,252,253,26,254],"农业人工智能","智能灌溉","温室种植",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[257],{"code":196,"slug":258,"name":259,"path":260,"parent_code":261},"irrigation","水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",{"status":39,"locked":14,"version":40,"method":41,"evidence":263},{"candidates":264,"agricultural_evidence":47},[265,270,275,279,286,292,298],{"code":196,"score":19,"supported":47,"title":266,"abstract":267,"tags":269},[258],[268,199,258],"灌溉",[253],{"code":32,"score":271,"supported":47,"title":272,"abstract":273,"tags":274},15,[119],[56,119],[254],{"code":45,"score":89,"supported":47,"title":276,"abstract":277,"tags":278},[],[26,125],[95,26],{"code":280,"score":77,"supported":47,"title":281,"abstract":282,"tags":285},"H2",[],[283,284],"土壤","soil",[],{"code":287,"score":60,"supported":47,"title":288,"abstract":289,"tags":291},"B1",[],[290],"番茄",[],{"code":293,"score":60,"supported":47,"title":294,"abstract":295,"tags":297},"H7",[],[296],"气候变化",[],{"code":66,"score":21,"supported":14,"title":299,"abstract":300,"tags":301},[],[],[252],[303,304],"MLP-AFLC 温室灌溉","神经网络 模糊控制 灌溉","MLP-AFLC温室灌溉-4371","10.1016\u002Fj.compeleceng.2026.111587",{"doi":306,"openalex_id":308,"authors":309,"venue":243,"cited_by_count":131,"oa_url":240,"card":324,"direction":329,"ingested_from":153},"W7220746994",[310,313,316,318,321],{"name":311,"orcid":312},"Jamel Riahi","https:\u002F\u002Forcid.org\u002F0009-0006-1434-4367",{"name":314,"orcid":315},"Hamza Nasri","https:\u002F\u002Forcid.org\u002F0009-0008-7343-4199",{"name":317,"orcid":8},"Abdelkader Mami",{"name":319,"orcid":320},"Mario Carpentieri","https:\u002F\u002Forcid.org\u002F0000-0001-5165-5873",{"name":322,"orcid":323},"Silvano Vergura","https:\u002F\u002Forcid.org\u002F0000-0002-0826-9237",{"tldr":325,"method":326,"finding":327,"direction":151,"opportunity":328},"提出MLP-ANN预测温室微气候并驱动自适应模糊控制器，实现预测性智能灌溉。","MLP神经网络预测温湿度与土壤水分，结合自适应模糊逻辑控制流量。","预测R²达0.999，模糊控制器动态调节0-18 LPH流量，显著节水节能。","可探索多作物多环境迁移学习与边缘部署，降低预测模型算力需求。","农业人工智能与决策模型","2026-10-07T23:30:43.848308Z",{"id":332,"title":333,"url":334,"summary":335,"summary_zh":336,"content":8,"source_name":337,"source_url":334,"published_at":338,"category":12,"cover_url":8,"hotness":339,"is_selected":14,"score":340,"score_detail":341,"sources":343,"tags":347,"primary_topic":350,"secondary_topics":351,"classification":352,"search_phrases":381,"slug":384,"view_count":21,"doi":385,"paper":386,"created_at":396},3821,"Transforming Agriculture Through Technology And Smart Farming Innovations","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23036819","Abstract In the contemporary agricultural landscape, a significant paradigm shift is taking place as farmers across the globe increasingly adopt advanced technologies and innovative farming practices. This transformation highlights the profound impact of modern agricultural technologies and the key trends shaping the future of food production. A central component of this revolution is the integration of precision agriculture, wherein farmers utilize cutting-edge tools such as GPS systems, sensors, and drones to optimize resource allocation. This approach enhances crop management efficiency, reduces resource wastage, and improves overall productivity. The rise of smart farming has introduced data-driven decision-making processes, enabling farmers to monitor, analyze, and manage crops with remarkable accuracy. In addition, advancements in biotechnology and genetic engineering have facilitated the development of genetically modified crops with improved resilience, enhanced nutritional value, and greater resistance to pests and diseases. These innovations not only increase agricultural output but also promote sustainability and environmentally responsible farming practices. Moreover, the emergence of vertical farming and controlled-environment agriculture marks a significant departure from conventional farming methods. By employing hydroponic and aeroponic systems in vertically structured environments, crops can be cultivated efficiently with minimal land use and optimized resource consumption. This method is particularly advantageous in urban areas where arable land is limited, offering a sustainable solution to growing food demands. Furthermore, robotics and automation are reshaping agricultural operations. Technologies such as autonomous tractors, robotic harvesters, and drone-assisted planting systems are reducing reliance on manual labor while enhancing operational efficiency. These advancements help address labor shortages and strengthen the overall resilience and sustainability of the agricultural sector.","在当代农业格局中，随着全球农民日益采用先进技术和创新耕作方式，正在发生重大的范式转变。这一变革凸显了现代农业技术的深远影响以及塑造未来粮食生产的关键趋势。这场革命的核心组成部分是精准农业的整合，农民利用GPS系统、传感器和无人机等前沿工具来优化资源配置。这种方法提高了作物管理效率，减少了资源浪费，并提升了整体生产力。智慧农业的兴起引入了数据驱动的决策过程，使农民能够以极高的精度监测、分析和管理作物。此外，生物技术和基因工程的进步促进了转基因作物的开发，这些作物具有更强的适应力、更高的营养价值和更强的抗病虫害能力。这些创新不仅提高了农业产量，还促进了可持续性和环境友好的耕作方式。此外，垂直农业和受控环境农业的出现标志着对传统耕作方式的重大突破。通过在垂直结构中采用水培和气培系统，作物可以在最少的土地利用和优化的资源消耗下高效种植。这种方法在城市地区尤为有利，因为那里可耕地有限，为日益增长的粮食需求提供了可持续的解决方案。此外，机器人技术和自动化正在重塑农业作业。自主拖拉机、机器人收割机和无人机辅助种植系统等技术正在减少对人工劳动的依赖，同时提高作业效率。这些进步有助于应对劳动力短缺问题，并增强农业部门的整体韧性和可持续性。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-30T00:00:00Z",25,65,{"impact":19,"substance":117,"depth":17,"authority":17,"freshness":20,"relevant":21,"comment":342},"Zenodo 平台上的智慧农业综述性论文，覆盖面广但缺乏新数据与具体案例，可作为主题聚合素材而非每日精选。",[344,345],{"name":337,"url":334},{"name":337,"url":346},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23036818",[95,348,349,26,171],"农业机器人","生物育种",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":353},{"candidates":354,"agricultural_evidence":47},[355,360,365,371,376],{"code":45,"score":356,"supported":47,"title":357,"abstract":358,"tags":359},11,[],[95,26,125],[95,26],{"code":32,"score":46,"supported":47,"title":361,"abstract":362,"tags":364},[],[171,363],"hydroponic",[171],{"code":366,"score":60,"supported":47,"title":367,"abstract":368,"tags":370},"F5",[],[369],"收割机",[],{"code":372,"score":21,"supported":14,"title":373,"abstract":374,"tags":375},"F2",[],[],[349],{"code":377,"score":21,"supported":14,"title":378,"abstract":379,"tags":380},"F6",[],[],[348],[382,383],"农业机器人 垂直农业 智慧农业 生物育种","农业机器人 垂直农业","农业机器人垂直农业智慧农业生物育种-3821","10.5281\u002Fzenodo.23036819",{"doi":385,"openalex_id":387,"authors":388,"venue":337,"cited_by_count":131,"oa_url":334,"card":391,"direction":151,"ingested_from":153},"W7214716174",[389],{"name":390,"orcid":8},"R.L. Poonguzhali",{"tldr":392,"method":393,"finding":394,"direction":151,"opportunity":395},"综述精准农业、智慧农业、生物技术、垂直农业与机器人自动化如何重塑未来粮食生产。","文献综述，整合GPS、传感器、无人机、基因工程、水培气培及自动化技术案例。","技术融合可提升效率、减少浪费、促进可持续，并应对劳动力短缺与耕地限制。","可量化评估垂直农业与机器人系统在中小农场的经济可行性及碳足迹，填补实证空白。","2026-09-30T23:30:09.413038Z",{"id":398,"title":399,"url":400,"summary":401,"summary_zh":8,"content":402,"source_name":403,"source_url":8,"published_at":404,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":405,"score_detail":406,"sources":408,"tags":410,"primary_topic":416,"secondary_topics":417,"classification":419,"search_phrases":442,"slug":445,"view_count":60,"doi":8,"paper":8,"created_at":446},1034,"AI-Enabled IoT-Based Hydroponic Farming with Embedded Automation and Nutrient Prediction","https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446","原文标题:AI-Enabled IoT-Based Hydroponic Farming with Embedded Automation and Nutrient Prediction。作者:Jehangir Arshad、Fawad Azeem、Ayesha Butt、Maha Chaudhary、Rana Saad Safdar、M. Kamran Joyo、Izanoordina Ahmad、Prajoona Valsalan、Husham M. Ahmed(COMSATS University Islamabad等)。DOI: 10.3390\u002Ffi18090446。论文构建一个由Raspberry Pi控制、完全监测并持续评估的水培农业模拟器,采用多种传感器、数据管理与处理以及自动化环境调节。该智能水培系统采用双模型机器学习流水线:一个通过图像分析识别植物病害,另一个基于传感器数据评估植物营养水平。两个模型的数据通过云端数字孪生(DT)合并,实现远程访问并提供闭环控制,用于水培环境中的灌溉、营养剂量和所有环境因子管理。该研究展示了将物联网传感、机器学习和数字孪生模拟组合成可运行超物理系统来开发可扩展、数据驱动的精准农业解决方案的能力。","## 1. Introduction and Background\n\nControlled-environment agriculture is moving quickly toward distributed sensing, edge computing, and cloud analytics to reduce manual supervision. However, in practice, many hydroponic platforms treat sensing, vision-based plant health screening, cloud monitoring, and actuator control as separate modules that happen to sit next to each other rather than work together. This separation leaves a gap between collecting real-time data and actually closing the loop on decisions. This work addresses this gap by integrating multi-sensor hydroponic monitoring, RGB image-based disease screening, nutrient-status prediction, actuator control, and Digital Twin visualization within a single operational platform.\n\nMeeting the nutritional needs of a growing global population is hardly easy, and the challenges are not helped by climate change, dwindling freshwater resources, or the gradual loss of arable land [[1](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B1-futureinternet-18-00446),[2](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B2-futureinternet-18-00446)]. In urban areas, where space is limited and supply chains are extended, traditional farming has difficulty meeting demand. The solution that has emerged as more and more viable is controlled-environment agriculture (CEA): growers can control temperature, humidity, light, and nutrients, and thus grow food reliably and efficiently, irrespective of the outside climate [[3](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B3-futureinternet-18-00446),[4](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B4-futureinternet-18-00446)]. Hydroponics sits at the center of CEA. Growing plants without soil reduces water consumption significantly and often yields better per square meter than conventional planting. But the problem is that most hydroponic systems continue using manual monitoring and static control rules, causing the systems to be slow to change when problems start [[5](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B5-futureinternet-18-00446),[6](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B6-futureinternet-18-00446)]. With advancements in recent developments in the fields of IoT, AI, and cloud computing, it is now possible to develop systems that monitor themselves, predict issues, and intervene without human supervision [[7](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B7-futureinternet-18-00446),[8](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B8-futureinternet-18-00446),[9](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B9-futureinternet-18-00446)].\n\nSensors for pH, electrical conductivity (EC), total dissolved solids (TDS), ultrasonic sensors, temperature, and humidity are already commercially available. Their data can be used to feed into machine learning models, enabling early detection of nutrient imbalances, optimization of fertigation timings, and identification of environmental anomalies [[10](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B10-futureinternet-18-00446),[11](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B11-futureinternet-18-00446)]. On the vision side, deep learning has enabled rapid and accurate automatic disease diagnosis of leaf images [[12](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B12-futureinternet-18-00446),[13](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B13-futureinternet-18-00446),[14](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B14-futureinternet-18-00446)]. The Digital Twin adds to this by keeping a current virtual model of the physical system. The real environment is simulated, and operators can run trials on a virtual plant prior to implementation on their own crop [[15](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B15-futureinternet-18-00446),[16](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B16-futureinternet-18-00446),[17](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B17-futureinternet-18-00446)]. All these developments at the individual level, however, do not make a system that integrates sensing, prediction, visualization, and actuation under a single roof commonplace [[4](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B4-futureinternet-18-00446),[18](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B18-futureinternet-18-00446)]. This paper introduces Hydro Grow, a hydroponic farming system, which combines two different types of AI models, IoT sensors, and an Azure-supported Digital Twin in a functional cyber-physical system. The details of the specific contribution are given below.\n\n*   A dual-stream AI pipeline with computer vision and inference extracted from the sensors.\n\n*   MobileNetV2 for plant disease classification and a decision tree for nutrient deficiency prediction.\n\n*   Cloud-synchronized digital twin with Microsoft Azure Digital Twins for real-time monitoring.\n\n*   Closed-loop actuator control (pumps, dozers, ventilation) with all control based on sensor data.\n\n*   An inbuilt AI chatbot for handling client queries was integrated with a web-based monitoring dashboard.\n\nMost early hydroponic systems were manually operated. Nutrient and pH levels had to be manually analyzed and adjusted, and early automated systems used mechanical timers and simple pumps. These instruments made the work easier but did not add any intelligence to the process, as they operated on a timetable regardless of the conditions in the reservoir. Microcontroller-based systems allow for simple feedback, irrigation, and nutrient timing. They are less reliable than timers but still fundamentally reactive; they might be able to obtain a signal when a threshold is crossed, but they would not be able to predict when a crossing is likely to occur or why it is occurring. This resulted in their being sufficient only for small, stable systems and unsuitable for more complex systems [[19](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B19-futureinternet-18-00446)]. IoT has transformed the landscape by making it possible and affordable to continuously monitor multiple parameters [[1](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B1-futureinternet-18-00446),[20](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B20-futureinternet-18-00446)]. Today, grain growers have far greater real-time access to information about what is happening with their distributed sensor networks. Thus, there is another big space—intelligent interpretation of that data, and that is where the technology lies. Most deployed systems are still rule-based (hard-coded) and have not yet learned models, which means that they cannot adapt to unknown environments or optimize their behavior over time [[21](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B21-futureinternet-18-00446),[22](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B22-futureinternet-18-00446)].\n\nThe remainder of this paper is organized as follows: [Section 2](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec2-futureinternet-18-00446) reviews the relevant prior work. [Section 3](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec3-futureinternet-18-00446) covers system modeling and simulation. [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec4-futureinternet-18-00446) discusses the implementation of hardware and software. [Section 5](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec5-futureinternet-18-00446) presents the evaluation of the model’s performance. [Section 6](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec6-futureinternet-18-00446) provides a comparison of Hydro Grow with other systems. [Section 7](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec7-futureinternet-18-00446) represents the results and discussion, the conclusion is provided in [Section 8](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#sec8-futureinternet-18-00446).\n\n## 2. Related Work\n\nIoT-based monitoring has become standard in commercial and research hydroponic systems, and monitoring using the Internet of Things (IoT) is becoming commonplace. Platforms based on ESP32 microcontroller electronics have been shown to provide well-defined advantages for nutrient management and decrease manual nutrient monitoring [[9](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B9-futureinternet-18-00446),[23](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B23-futureinternet-18-00446),[24](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B24-futureinternet-18-00446),[25](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B25-futureinternet-18-00446)]. These systems have great functionality within their design specifications but come to a halt when the conditions change outside of the designer’s expectations [[6](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B6-futureinternet-18-00446),[21](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B21-futureinternet-18-00446)]. Machine learning has been used in hydroponics primarily for predictive operations such as predicting water demand, nutrient uptake, and alerting if the operations fall out of the range [[10](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B10-futureinternet-18-00446),[22](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B22-futureinternet-18-00446),[26](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B26-futureinternet-18-00446)]. Structured sensor data have been used to demonstrate a high level of accuracy using decision trees and other hybrid supervised methods [[10](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B10-futureinternet-18-00446),[11](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B11-futureinternet-18-00446)]. The problem, however, is that most of these models are trained separately and then applied. They are not updated as the crop and environment change and do not provide feedback to actuation systems [[27](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B27-futureinternet-18-00446)]. Plant health assessment using computer vision has come a long way. CNN architectures are well-suited to reliably classify diseases from leaf images, for example, MobileNet variants, ResNet, and other lightweight architectures, and transfer learning from ImageNet weights significantly reduces the amount of data required [[14](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B14-futureinternet-18-00446),[28](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B28-futureinternet-18-00446),[29](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B29-futureinternet-18-00446),[30](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B30-futureinternet-18-00446),[31](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B31-futureinternet-18-00446)]. However, there are complications when using the vision system in the real world; lighting, leaf overlap, and other background clutter can all reduce accuracy in field conditions, which is why the use of sensor-based confirmation makes a good design choice. [Table 1](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#table_body_display_futureinternet-18-00446-t001) summarizes the gaps identified in each of these research threads.\n\n**Table 1.** Research gaps identified in the related literature.\n\nThe literature has considered digital twin applications in hydroponic and aquaponic systems, with some platforms moving towards complete robot integration, such as DigiHortiRobot [[16](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B16-futureinternet-18-00446),[17](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B17-futureinternet-18-00446),[32](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B32-futureinternet-18-00446),[33](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B33-futureinternet-18-00446)]. It is still a challenge to achieve real-time synchronization between physical and virtual environments, and the communication overhead can be substantial. Nevertheless, the idea is intriguing: having a well-maintained twin system that allows the operator to test interventions in a virtual environment before deploying them in the actual system.\n\n### 2.1. Smart and Automated Hydroponic Systems\n\nTransitioning from manual scheduling to sensor-based control has been a gradual process. Most early systems used presets without feedback, and later started to implement feedback based on sensor data using ESP32 microcontrollers [[19](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B19-futureinternet-18-00446),[24](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B24-futureinternet-18-00446)]. Despite this, many implementations rely on hard-coded thresholds and fall short in regions where the environmental conditions fall outside the threshold ranges for which they are calibrated [[21](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B21-futureinternet-18-00446),[25](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B25-futureinternet-18-00446)]. Awal et al. [[23](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B23-futureinternet-18-00446)] created a house-type hydroponic system for Bangladesh that measures various parameters in real-time without making any prediction, nor do they integrate it with a cloud.\n\n### 2.2. AI and Machine Learning for Hydroponics\n\nNutrient concentration prediction and finding optimum growth conditions have been predicted with reasonable success using random forest, linear regression, and hybrid AI methods [[11](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B11-futureinternet-18-00446),[26](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B26-futureinternet-18-00446)]. Rahman et al. [[26](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B26-futureinternet-18-00446)] developed an IoT system that integrates crop recommendation and nutrient monitoring, demonstrating the potential of IoT systems when using real sensor data as input for ML. The models adopted in the vision part are MobileNet-family models and VGG variants, which have been proven to be well suited for edge deployment due to their small size and efficiency [[14](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B14-futureinternet-18-00446),[28](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B28-futureinternet-18-00446),[30](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B30-futureinternet-18-00446)]. Labeled hydroponic datasets are small, and transfer learning from ImageNet can be used in these cases [[29](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B29-futureinternet-18-00446),[31](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B31-futureinternet-18-00446)]. The common challenge in all this work is how to link these models to the actuation layer. Most systems produce a prediction and then call it a day and leave it up to the human to decide on the next action to take [[22](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B22-futureinternet-18-00446),[27](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B27-futureinternet-18-00446)]. Recent agricultural vision research is moving beyond image-level classification toward fine-grained segmentation and multimodal reasoning; Yang et al. proposed PestScope, an exclusion-aware large multimodal model for fine-grained agricultural pest segmentation, showing that agricultural diagnosis benefits from models that can localize visually similar stress or pest patterns [[34](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B34-futureinternet-18-00446)]. The present work instead adopts MobileNetV2 for lightweight image-level disease screening on the Raspberry Pi 4B, prioritizing edge deployability over pixel-level segmentation. Feature-fusion strategies have also shown value in other sensor-driven automation tasks: Wang et al. improved seam extraction for teachless welding robots using a multichannel feature-fusion network, demonstrating that combining complementary visual features can improve perception reliability under challenging imaging conditions [[35](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B35-futureinternet-18-00446)]. This supports the design choice of combining visual plant-health screening with sensor-based nutrient inference rather than relying on a single sensing stream.\n\nTaken together, the shortfall in the literature is not a lack of sensors or models. Recent hydroponic systems increasingly combine IoT monitoring with cloud dashboards; however, most still rely on threshold-based actuation or evaluate machine learning models separately from the control layer. What is missing is an experimental validation of a unified architecture in which sensing, edge inference, cloud synchronization, visualization, and actuator response are tested together as one closed-loop system, rather than as components verified in isolation. Hybrid CNN-Transformer architectures have also been applied to high-resolution agricultural remote-sensing tasks; for example, Xie et al. used a boundary-guided CNN-Transformer network for cropland parcel mapping, reflecting the wider movement toward hybrid visual architectures in agricultural image interpretation [[36](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B36-futureinternet-18-00446)]. However, such models are computationally heavier than MobileNetV2 and less suitable for the Raspberry Pi-based edge deployment targeted in this work.\n\n## 3. Materials and Methods\n\n### 3.1. Conceptual Framework\n\nIn this system, the physical growing system employed is deep water culture (DWC), in which the root system hangs in the solution, and the solution is supplied with oxygen and nutrients [[1](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B1-futureinternet-18-00446),[5](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B5-futureinternet-18-00446)]. Water moves and is oxygenated by an air pump, and stagnant water results in anaerobic conditions that adversely affect the roots and yield. DWC is a suitable application for an automated system because of the relatively low changes in the system chemistry, allowing the sensor to settle before the actuator reacts to the change in the water quality. It consists of five layers: real-time IoT monitoring, edge AI inference, cloud-based analytics, digital twin synchronization, and a human-supervised web interface [[7](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B7-futureinternet-18-00446),[15](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B15-futureinternet-18-00446),[17](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B17-futureinternet-18-00446)]. These layers are loosely coupled, such that a drop to the cloud does not eliminate the use of local sensing or edge inference. Five sensors were used to measure the important environmental variables. The pH sensor determines the acidity of the nutrient solution, which directly influences mineral absorption [[10](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B10-futureinternet-18-00446),[25](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B25-futureinternet-18-00446)]. The total dissolved mineral concentration was measured using an EC\u002FTDS sensor. The DHT22 is a sensor that measures the temperature and air humidity around the plant canopy of a tree. The water level of the reservoir was monitored using an ultrasonic sensor. Images of the leaves were captured using a Raspberry Pi Camera Module V2 for visual disease screening [[12](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B12-futureinternet-18-00446),[14](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B14-futureinternet-18-00446)].\n\n### 3.2. Four-Layer System Architecture\n\nThe main architecture is a four-layer pyramid to separate sensing, edge control, cloud services, and user-facing applications, as presented in [Figure 1](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#fig_body_display_futureinternet-18-00446-f001). These concerns should be kept separate to allow the system to be debugged, updated, and scaled without redesign [[7](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B7-futureinternet-18-00446),[8](https:\u002F\u002Fwww.mdpi.com\u002F1999-5903\u002F18\u002F9\u002F446#B8-futureinternet-18-00446)]. The sensing layer is a data-collection layer. Analog sensors are connected to the ADC circuits on the Raspberry Pi 4B, and digital sensors are connected through I2C or GPIO. No analysis is performed here.\n\n**Figure 1.** Four-layered architecture of the proposed AI-enhanced intelligent hydroponic farming.\n\nThe control layer is installed on the Raspberry Pi 4B. It receives sensor data, checks against target ranges, and activates the relay-controlled actuators, which include the water pump, nutrient dispensers, and LED grow lights when deviations appear. The system includes the sensing layer (pH, EC, and ultrasonic imaging), control layer (logical controller, microcontroller, supervisory control, and AI controller), processing layer (cloud server, AI services, AI services control, and artificial intelligence unit), and user layer (mobile application). Microsoft Azure is used in the c","MDPI Future Internet 2026, 18(9): 446","2026-08-23T16:00:00Z",69,{"impact":271,"substance":19,"depth":165,"authority":166,"freshness":77,"relevant":21,"comment":407},"论文提出整合多传感器、视觉检测、营养预测与数字孪生的水培系统，方法新颖，数据详实，对智慧农业有参考价值，但影响范围限于细分领域。",[409],{"name":403,"url":400},[411,412,50,26,413,414,415],"农业物联网","机器学习","智能水培","MobileNetv2","Future Internet",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[418],{"code":102,"slug":103,"name":104,"path":105,"parent_code":106},{"status":39,"locked":14,"version":40,"method":41,"evidence":420},{"candidates":421,"agricultural_evidence":47},[422,426,430,434,438],{"code":102,"score":271,"supported":47,"title":423,"abstract":424,"tags":425},[112],[96,112],[411],{"code":32,"score":356,"supported":47,"title":427,"abstract":428,"tags":429},[363],[363],[],{"code":45,"score":89,"supported":47,"title":431,"abstract":432,"tags":433},[],[26,50],[50,26],{"code":66,"score":53,"supported":47,"title":435,"abstract":436,"tags":437},[],[412],[412],{"code":196,"score":60,"supported":47,"title":439,"abstract":440,"tags":441},[],[268],[],[443,444],"农业物联网 数字孪生 智能水培 机器学习","农业物联网 数字孪生","农业物联网数字孪生智能水培机器学习-1034","2026-08-28T00:07:02.219053Z",{"id":448,"title":449,"url":450,"summary":451,"summary_zh":8,"content":452,"source_name":453,"source_url":8,"published_at":454,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":455,"score_detail":456,"sources":458,"tags":460,"primary_topic":464,"secondary_topics":468,"classification":474,"search_phrases":490,"slug":493,"view_count":60,"doi":8,"paper":8,"created_at":494},379,"TMAF-YOLO:温室番茄成熟度与缺陷果实轻量化实时检测模型","https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950","中国农业大学黄晨晓、何林然、黄文涛、张晓绚等于Sensors发表TMAF-YOLO模型——基于YOLOv8n的轻量化番茄成熟度与缺陷果实检测模型,隶属中国农业大学120周年校庆专刊。该模型以LGhostConv替换选定卷积结构减少冗余计算;引入番茄成熟度感知聚合融合模块TMAF增强检测分支的颜色、纹理和局部结构特征表示能力;使用MA-CB Focal Loss改善类别不平衡和难样本条件下的学习。实验结果显示,Precision、Recall、mAP50和mAP50-95分别达到0.889、0.873、0.954和0.776,模型仅含2.647M参数,推理速度209.030 FPS,优于YOLOv11n和YOLOv12n,同时保持实时推理性能。","## 1. Introduction\n\nTomato is an important vegetable crop in protected agriculture. Its maturity directly affects harvesting time, transport management, postharvest ripening, quality grading, and commercial value. In greenhouse environments, tomato fruits are often affected by leaf occlusion, fruit overlap, uneven illumination, and scale variation. Manual maturity assessment is inefficient and easily influenced by subjective experience. Therefore, a rapid, stable, and non-contact detection method is needed to support automated harvesting, grading management, and robotic visual perception in protected agriculture [[1](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B1-sensors-26-04950),[2](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B2-sensors-26-04950),[3](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B3-sensors-26-04950)].\n\nAmong deep learning detection methods, region-based convolutional neural network (R-CNN) models and their variants were among the earlier methods applied to fruit and vegetable detection and counting tasks. An improved Faster R-CNN model, MatDet, was proposed for multi-object tomato maturity detection in complex scenarios [[1](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B1-sensors-26-04950)]. The model used a 50-layer residual network (ResNet-50), region-of-interest alignment (RoIAlign), and a path aggregation network (PANet) to improve detection under occlusion, overlap, and illumination variation, achieving a mean average precision (mAP) of 96.14%. Faster R-CNN has also been used for tomato-related detection tasks. Highly occluded immature tomatoes on plants were detected using Faster R-CNN, with an average precision of 87.83% [[4](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B4-sensors-26-04950)]. Greenhouse tomato flowers were detected and counted using Faster R-CNN, achieving an average precision of 96.02% [[5](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B5-sensors-26-04950)]. These methods have strong localization ability and can provide stable target region information. However, their model structures are relatively complex, and their inference speed and lightweight deployment ability remain limited.\n\nWith the development of Transformer architectures, end-to-end detectors such as Detection Transformer (DETR), DETR with Improved deNoising anchOr boxes (DINO), and Real-Time Detection Transformer (RT-DETR) have been introduced into object detection tasks. DETR formulates object detection as a set prediction problem [[6](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B6-sensors-26-04950)]. DINO improves the convergence and detection performance of DETR-like models through denoising training and query optimization [[7](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B7-sensors-26-04950)]. RT-DETR further improves the Transformer detection structure for real-time detection requirements [[8](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B8-sensors-26-04950)]. In fruit and vegetable maturity recognition, PDSI-RTDETR was proposed for tomato maturity detection. Compared with the original RT-DETR, its mAP50 and mAP50-95 increased by 3.9 and 4.1 percentage points, respectively [[9](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B9-sensors-26-04950)]. An improved RT-DETR was also applied to fruit ripeness detection, indicating the potential of Transformer detectors in global feature modeling [[10](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B10-sensors-26-04950)]. However, these models usually have higher training costs, larger parameter sizes, and greater hardware requirements.\n\nIn contrast, You Only Look Once (YOLO) series models provide a better balance between detection accuracy and inference speed. They have become common methods for fruit and vegetable detection and maturity recognition. A lightweight improved YOLOv5 model was proposed for real-time tomato detection and mobile deployment [[2](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B2-sensors-26-04950)]. Compared with the original YOLOv5s, the number of parameters and floating-point operations (FLOPs) were reduced by 78% and 84.15%, respectively, while maintaining high detection accuracy. RGB-D information was combined with YOLOv5 to detect and count greenhouse tomato clusters [[3](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B3-sensors-26-04950)]. MTD-YOLO was proposed for multitask detection of cherry tomato fruits, clusters, and maturity [[11](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B11-sensors-26-04950)]. The model achieved an overall detection score of 86.6% and an average inference time of 4.9 ms. An improved YOLOX model was proposed for tomato ripeness and stem recognition, achieving an mAP of 92.17% and improving recognition under sample imbalance and scale variation [[12](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B12-sensors-26-04950)]. YOLOv8 was combined with a lightweight Swin Transformer for strawberry ripeness detection [[13](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B13-sensors-26-04950)]. Compared with YOLOv5s, CenterNet, and the Single Shot MultiBox Detector (SSD), the mAP50 increased by 1.6%, 33.5%, and 3.4%, respectively. These results indicate that improved YOLO models are suitable for fine-grained fruit maturity recognition. In addition, recent YOLO-based studies on tomato detection and ripening recognition have improved detection performance using attention mechanisms, lightweight structures, multiscale feature fusion, and postprocessing optimization [[14](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B14-sensors-26-04950),[15](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B15-sensors-26-04950),[16](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B16-sensors-26-04950),[17](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B17-sensors-26-04950)].\n\nExisting studies have verified the effectiveness of deep learning methods in fruit and vegetable detection and maturity recognition. However, further improvement is still needed for in situ greenhouse tomato detection. First, the color transition between semi-mature and mature tomatoes is continuous, making fine-grained classification difficult. Second, leaf occlusion, fruit overlap, and illumination variation weaken maturity-related feature representation. Third, defective samples are relatively limited, and conventional classification losses may underemphasize minority and hard samples. Previous studies have shown that Focal Loss and Class-Balanced Loss can adjust the loss weights of hard and imbalanced samples, thereby improving the learning ability for long-tailed classes and difficult samples [[18](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B18-sensors-26-04950),[19](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B19-sensors-26-04950)]. To address these issues, this study proposes TMAF-YOLO based on YOLOv8n.\n\nThe main contributions of this study are as follows.\n\n(1)\nA lightweight tomato maturity detection model was constructed for complex greenhouse scenarios. Inspired by the low-cost feature generation strategy in GhostNet, an LGhostConv module was designed and embedded in key positions of the Backbone and Neck [[20](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#B20-sensors-26-04950)]. This design reduces the number of parameters and computational cost while maintaining detection performance.\n\n(2)\nA tomato maturity-aware aggregation fusion module, termed TMAF, was proposed. This module was designed to address continuous color transitions between adjacent maturity stages, weak local texture differences, and insufficient feature representation caused by occlusion and overlap. By integrating local texture enhancement and channel-spatial recalibration, TMAF strengthens maturity-related feature representation.\n\n(3)\nMA-CB Focal Loss was developed to improve learning under class imbalance and hard-sample conditions. By combining class-balanced weights with the Focal Loss modulation term, the loss enhances learning for minority and difficult samples, including defective fruits and adjacent maturity stages.\n\n## 2. Materials and Methods\n\n### 2.1. Overall Method Design\n\nThis study focuses on in situ tomato maturity detection before greenhouse harvesting. An overall application workflow was constructed, including greenhouse image acquisition, fruit detection and category recognition, harvesting decision-making, and postharvest grading management. As shown in [Figure 1](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#fig_body_display_sensors-26-04950-f001), tomato images were first acquired under natural growth conditions in the greenhouse. Then, the developed detection model was used to identify fruit locations, maturity stages, and defective fruits. Finally, the detection results were used to support harvesting time determination, defective fruit screening, and postharvest grading and quality management. It should be noted that this study mainly focuses on greenhouse tomato maturity and defective fruit detection. The harvesting, grading, and management processes in the figure are used to illustrate the potential application scenarios of the proposed model.\n\n**Figure 1.** Overall workflow of greenhouse tomato maturity detection.\n\n### 2.2. Dataset Construction and Annotation\n\nThe self-collected images used in this study were acquired in a tomato greenhouse at the Tianjin Academy of Agricultural Sciences, China, from 10 to 17 January 2026. Image acquisition was conducted between 9:00 a.m. and 12:00 p.m. using a HUAWEI Mate 60 smartphone (Huawei Device Co., Ltd., Dongguan, China) in automatic shooting mode. Exposure, focus, ISO sensitivity, and shutter speed were adjusted automatically by the built-in camera system. The shooting distance ranged from approximately 0.5 to 2.0 m, and the image resolution was 1280 × 760 pixels. Images were captured under naturally varying greenhouse illumination; illuminance was not measured instrumentally.\n\nTo represent practical greenhouse conditions, images were acquired from different viewing angles and distances. The collected scenes included fruits at different maturity stages and scales, together with leaf and branch occlusion, fruit overlap, dense fruit distribution, small targets, complex backgrounds, and local image blur. Representative examples of self-collected images are shown in [Figure 2](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#fig_body_display_sensors-26-04950-f002).\n\n**Figure 2.** Representative tomato images acquired under typical greenhouse conditions: (**a**) mature fruits; (**b**) adjacent fruits at different maturity stages; (**c**) fruit overlap and foliage interference; (**d**) leaf and branch occlusion; (**e**) small targets in a relatively long-range view; and (**f**) local image blur.\n\nA total of 596 valid images were collected from the greenhouse. Because defective fruits were relatively scarce, 33 additional defective-fruit images were selected from the Tomato Fruit Diseases Dataset available on Kaggle ([https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fprofnourasemary\u002Ftomato-dataset](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fprofnourasemary\u002Ftomato-dataset), accessed on 1 July 2026). The additional images mainly consisted of tomato fruits exhibiting visible disease symptoms, together with a smaller number showing physical damage, decay, cracking, or other abnormal surface characteristics. The selected public images were annotated according to the same criteria as the self-collected images.\n\nThe 596 self-collected images were divided at the original-image level into training, validation, and test subsets containing 470, 63, and 63 images, respectively. The 33 public images were added exclusively to the training set to increase the number and diversity of defective-fruit samples. Consequently, the final training, validation, and test sets contained 503, 63, and 63 images, respectively. The validation and test sets therefore consisted entirely of self-collected greenhouse images.\n\nThe annotated tomato instances were divided into four categories: immature, semi-mature, mature, and defective. The first three categories represent successive normal maturity stages based primarily on fruit color and visible appearance. The defective category includes fruits with visible disease spots, physical damage, decay, cracking, or other abnormal surface characteristics and was treated as an independent appearance category rather than as a normal maturity stage.\n\nAll tomato instances were manually annotated with bounding boxes using LabelMe v6.0.0. The original annotations were stored in JSON format and subsequently converted into the format required by the YOLO detection framework. Partially occluded or locally blurred fruits were retained when their categories and principal visible boundaries could be reliably determined. Instances were excluded only when severe occlusion, insufficient visible information, or ambiguous appearance prevented consistent annotation. All annotations were reviewed according to uniform class definitions to ensure labeling consistency.\n\nThe final dataset contained 629 images and 2505 annotated tomato instances, including 1170 immature, 788 semi-mature, 304 mature, and 243 defective instances. Data augmentation was performed after dataset partitioning and was applied only to the training set. The validation and test sets were not augmented. The training set was used for model parameter optimization, the validation set for training monitoring and model selection, and the test set exclusively for final performance evaluation. The image and instance distributions before data augmentation are summarized in [Table 1](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#table_body_display_sensors-26-04950-t001).\n\n**Table 1.** Distribution of images and annotated tomato instances before data augmentation.\n\n### 2.3. Construction of the TMAF-YOLO Model\n\n#### 2.3.1. TMAF-YOLO: An Improved YOLOv8n Model\n\nThe proposed TMAF-YOLO uses YOLOv8n as the baseline model. It is improved from three aspects: lightweight feature extraction, maturity-related feature fusion, and classification loss optimization. The overall architecture is shown in [Figure 3](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#fig_body_display_sensors-26-04950-f003). After the original image is input into the network, multiscale features are first extracted by the Backbone, in which selected convolutional structures are replaced by LGhostConv. Then, features at different scales are fused in the Neck, and TMAF modules are added to the P3\u002F8 and P4\u002F16 branches. Finally, the model outputs prediction results through three detection heads at P3\u002F8, P4\u002F16, and P5\u002F32 scales. During training, the model is optimized using a joint loss function composed of WIoU, DFL, and MA-CB Focal Loss. The specific designs of LGhostConv, TMAF, and MA-CB Focal Loss are introduced in the following sections.\n\n**Figure 3.** Overall architecture of TMAF-YOLO.\n\n#### 2.3.2. Lightweight Feature Extraction Module: LGhostConv\n\nTo reduce redundant computation during feature extraction and multiscale feature fusion, a lightweight Ghost convolution module, termed LGhostConv, was introduced into selected downsampling layers of the backbone and neck. Similar to GhostConv, LGhostConv generates output features through primary and complementary feature branches. However, it replaces the dense spatial convolution in the primary branch with a depthwise-separable operation and employs directional depthwise convolutions in the complementary branch.\n\nGiven an input feature map $X$ with $C_{i n}$ channels, LGhostConv first applies a $k \\times k$ depthwise convolution with stride s for spatial feature extraction and downsampling, followed by a $1 \\times 1$ pointwise convolution for channel interaction. This process generates approximately half of the required output channels as the primary features. The remaining channels are generated by dividing the primary features into two groups and applying $1 \\times 3$ and $3 \\times 1$ depthwise convolutions to capture horizontal and vertical information, respectively. The directional features are then fused with the primary features through a parameter-free residual connection and concatenated to form the final output.\n\nIgnoring bias terms, batch-normalization parameters, and activation operations, the number of parameters in LGhostConv is expressed as follows:\n\n$$\nP_{L G h o s t C o n v} = k^{2} C_{i n} + C_{i n} m + 3 n\n$$\n\n(1)\n\nwhere $m$ = ⌈$C_{o u t}$\u002F2⌉ and $n$ = $C_{o u t}$ − $m$ denote the numbers of primary and complementary feature channels, respectively. In comparison, a standard $k \\times k$ convolution requires $k^{2} C_{i n} C_{o u t}$ parameters. Therefore, LGhostConv reduces model complexity by replacing dense spatial channel interactions with depthwise spatial filtering, pointwise channel projection, and low-cost directional feature generation. Its effects on detection accuracy, model size, computational complexity, and inference speed are evaluated in [Section 3.3](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#sec3dot3-sensors-26-04950).\n\n#### 2.3.3. Tomato Maturity-Aware Aggregation Fusion Module: TMAF\n\nTo strengthen the representation of maturity- and defect-related cues in the fused neck features, a tomato maturity-aware aggregation fusion module, termed TMAF, was designed. TMAF is embedded in the P3 and P4 feature fusion paths of the YOLOv8n neck, where relatively high-resolution features are retained for fine-grained maturity and defect detection. The original P5 path is preserved to limit the additional computational cost. As illustrated in [Figure 4](https:\u002F\u002Fwww.mdpi.com\u002F1424-8220\u002F26\u002F15\u002F4950#fig_body_display_sensors-26-04950-f004), TMAF comprises a lightweight texture aggregation branch and a TMCAttention submodule.\n\n**Figure 4.** Structure of the TMAF module.\n\nLet the input feature be $X$, with a size of $X \\in R^{B \\times C \\times H \\times W}$, where $B$, $C$, $H$, and $W$ denote the batch size, number of channels, feature map height, and feature map width, respectively. The input feature first passes through a $1 \\times 1$ convolution for channel mapping. It is then split along the channel dimension into a retained branch and a texture branch:\n\n$$\nX_{1} , X_{2} = S p l i t \\left(\\right. C o n v_{1 \\times 1} \\left(\\right. X \\left.\\right) \\left.\\right)\n$$\n\n(2)\n\nwhere $X_{1}$ is the retained branch, which maintains the direct transmission of input information. $X_{2}$ is the texture branch, which passes through three LTMBlocks in sequence to obtain texture-enhanced features at different levels, denoted as $T_{1}$, $T_{2}$, and $T_{3}$. Each LTMBlock consists of a $1 \\times 1$ convolution, a $3 \\times 3$ depthwise convolution, a $5 \\times 5$ depthwise convolution, and a $1 \\times 1$ convolution. The depthwise convolutions are used to supplement local texture and edge details. The $1 \\times 1$ convolutions are used for channel adjustment and feature fusion.\n\nThen, the retained branch $X_{1}$, and the texture branch outputs $T_{1}$, $T_{2}$, and $T_{3}$ are concatenated along the channel dimension. A $1 \\times 1$ convolution is then used to complete feature fusion and obtain the fused feature $F$:\n\n$$\n= C o n v_{1 \\times 1} \\left(\\right. C o n c a t \\left(\\right. X_{1} , T_{1} , T_{2} , T_{3} \\left.\\right) \\left.\\right)\n$$\n\n(3)\n\nTo further adjust maturity-related feature responses, TMCAttention is applied after feature fusion. This structure contains three branches: channel attention, spatial attention, and residual connection. Channel attention is used to adjust the importance of different feature channels. It helps the model focus on color, texture, and defect responses related to maturity discrimination. Spatial attention is used to highlight fruit region responses and reduce interference from irrelevant background features. The outputs of the two attention branches are denoted as $F_{c}$ and $F_{s}$, respectively.\n\nFinally, the outputs of the channel attention branch and the spatial attention branch are fused and added to the residual branch. The output feature of the TMAF module is obtained as follows:\n\n$$\nY = F + C o n v_{1 \\times 1} \\left(\\right. F_{c} + F_{s} \\left.\\righ","Sensors 2026-08-05","2026-08-05T00:00:00Z",62,{"impact":17,"substance":19,"depth":90,"authority":17,"freshness":53,"relevant":21,"comment":457},"针对温室番茄成熟度与缺陷检测的轻量化模型研究，方法有创新，但影响范围限于细分领域，时效性一般。",[459],{"name":453,"url":450},[461,462,27,463],"中国农业大学","Sensors","番茄成熟度",{"code":287,"slug":465,"name":466,"path":467,"parent_code":36},"vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",[469,473],{"code":66,"slug":470,"name":471,"path":472,"parent_code":106},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":475},{"candidates":476,"agricultural_evidence":47},[477,482,486],{"code":66,"score":17,"supported":47,"title":478,"abstract":480,"tags":481},[479],"YOLO",[479],[27],{"code":287,"score":356,"supported":47,"title":483,"abstract":484,"tags":485},[290],[290],[],{"code":32,"score":89,"supported":47,"title":487,"abstract":488,"tags":489},[56],[],[],[491,492],"中国农业大学 番茄成熟度 Sensors YOLOv8","中国农业大学 番茄成熟度","中国农业大学番茄成熟度SensorsYOLOv8-379","2026-08-10T23:54:46.339395Z"]