[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2153":3,"related-2153":70},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"primary_topic":9,"secondary_topics":24,"classification":25,"search_phrases":48,"slug":51,"view_count":52,"doi":53,"paper":54,"created_at":69},2153,"Design and Deployment of Multi-Class Image Classification Model on ESP32 Using a Lightweight CNN Approach","https:\u002F\u002Fdoi.org\u002F10.57041\u002Fp6f08552","Tremendous progress has been made in developing intelligent and autonomous systems in recent years, driven by AI software tools such as Chatbots and large language models (LLMs) for various applications. This has led to the adoption of smart technology solutions. Nonetheless, computer vision requires significant computational power, making it difficult to run on devices such as the ESP32. To showcase the development of smart technologies leveraging controller platforms, the research conducted training and deployment of an optimal multi-class image classification model on the ESP32 microcontroller platform, ensuring acceptable predictive performance across four classes: person, fruit, car, and unknown. It is important to clearly distinguish between the proposed system, which performs image classification and object detection. Regarding model training, the images were preprocessed to minimise computational cost. Nonetheless, the trained model achieved a peak validation accuracy of 99.34±1% when evaluated using five-fold cross-validation, with an average validation accuracy of 98±1% without altering the model or the training process, producing similar results. Moreover, TFLite Quantisation, a variant of the INT8 quantisation technique, was used to optimise the model size. The model was converted to C array format while retaining most features. To improve accuracy during training, the training dataset was augmented with large, high-quality images that had been filtered out. The chosen hardware used for this project was the ESP32-S3 N16R8 microcontroller with built-in OV2640 camera and the 3.2'' TFT to visualize the process of computer vision task execution in real time. The model works smoothly and predicts accurately, even with just a few samples per frame, to decrease response time latency. Successfully running the multi-class image classification model on the ESP32 microcontroller is the evidence of full independence of embedded vision solution without having anything to do with cloud technology.","近年来，在聊天机器人（Chatbots）和大语言模型（LLMs）等AI软件工具的推动下，智能自主系统的开发取得了巨大进展，并被应用于各种场景。这促进了智能技术解决方案的采用。然而，计算机视觉需要大量的计算能力，使其难以在ESP32等设备上运行。为了展示利用控制器平台开发智能技术的可行性，本研究在ESP32微控制器平台上训练并部署了一个最优多类图像分类模型，确保在四个类别（人、水果、汽车和未知）上具有可接受的预测性能。需要明确区分所提出的系统与执行图像分类和目标检测的系统。在模型训练方面，对图像进行了预处理以最小化计算成本。尽管如此，使用五折交叉验证评估时，训练模型达到了99.34±1%的最高验证准确率，在未改变模型或训练过程的情况下，平均验证准确率为98±1%，结果相似。此外，采用TFLite量化（TFLite Quantisation），即INT8量化技术的一种变体，来优化模型大小。模型被转换为C数组格式，同时保留了大部分特征。为了提高训练期间的准确率，训练数据集补充了此前被过滤掉的大尺寸高质量图像。本项目选用的硬件为内置OV2640摄像头的ESP32-S3 N16R8微控制器，以及3.2英寸TFT显示屏，用于实时可视化计算机视觉任务的执行过程。该模型运行流畅、预测准确，即使每帧仅有少量样本也能正常工作，从而降低响应时间延迟。在ESP32微控制器上成功运行多类图像分类模型，证明了嵌入式视觉解决方案的完全独立性，无需依赖云技术。",null,"International Journal of Emerging Engineering and Technology","2026-09-09T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"该论文聚焦ESP32微控制器上的轻量级CNN图像分类部署，属嵌入式计算机视觉通用技术研究，未涉及三农、农业信息化或智慧农业应用场景，与平台主题不相关。",[19],{"name":10,"url":6},[21,22,23],"农业人工智能","边缘计算","图像识别",[],{"status":26,"locked":14,"version":27,"method":28,"evidence":29},"needs_review","2026-10-22.2","rules",{"candidates":30,"agricultural_evidence":14,"reason":47},[31,41],{"code":32,"score":33,"supported":34,"title":35,"abstract":36,"tags":40},"I2",11,true,[],[37,38,39],"目标检测","计算机视觉","computer vision",[21,23],{"code":42,"score":43,"supported":14,"title":44,"abstract":45,"tags":46},"I5",1,[],[],[22],"农业相关性需要复核",[49,50],"农业人工智能 图像识别 边缘计算","农业人工智能 图像识别","农业人工智能图像识别边缘计算-2153",2,"10.57041\u002Fp6f08552",{"doi":53,"openalex_id":55,"authors":56,"venue":10,"cited_by_count":15,"oa_url":61,"card":62,"direction":66,"ingested_from":68},"W7212075138",[57,59],{"name":58,"orcid":9},"Abdul Hannan",{"name":60,"orcid":9},"Rao M. Asif","https:\u002F\u002Fgrsh.org\u002Fjournal1\u002Findex.php\u002Fijeet\u002Farticle\u002Fdownload\u002F202\u002F114",{"tldr":63,"method":64,"finding":65,"direction":66,"opportunity":67},"在ESP32微控制器上训练并部署轻量CNN多类图像分类模型，实现无云端的嵌入式视觉。","轻量CNN、TFLite INT8量化、五折交叉验证、ESP32-S3与OV26","模型验证准确率达99.34±1%，量化后仍保持性能，可在ESP32上实时准确分类。","智慧农业 \u002F 农业物联网","可探索将此类轻量模型用于田间作物\u002F病虫害实时识别，解决农业边缘设备算力与网络限制。","openalex","2026-09-11T23:30:14.512845Z",{"total":71,"page":43,"page_size":71,"items":72},6,[73,138,222,296,358,428],{"id":74,"title":75,"url":76,"summary":77,"summary_zh":9,"content":9,"source_name":78,"source_url":9,"published_at":79,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":80,"score_detail":81,"sources":88,"tags":90,"primary_topic":94,"secondary_topics":100,"classification":101,"search_phrases":124,"slug":127,"view_count":15,"doi":9,"paper":128,"created_at":137},5533,"基于低成本遥操作机械臂模仿学习的草莓果柄定点采摘","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.22765","该研究构建视觉运动模仿学习流程，统一感知、运动规划与接触操作，从人工遥操作演示训练基于ResNet-50骨干的动作分块Transformer策略，并在Jetson AGX Orin边缘平台以15赫兹自主部署。策略在130条演示、十万梯度步内损失由2.84降至0.091，数据集与模型权重已开源。预印本，已被IROS农业机器人研讨会接收。","arXiv 预印本 2026-10-16","2026-10-16T00:00:00Z",68,{"impact":82,"substance":83,"depth":84,"authority":85,"freshness":86,"relevant":43,"comment":87},16,21,18,9,4,"方法新颖、数据与开源材料扎实，但属预印本且已过一周，适合作为细分技术进展收录而非头条精选。",[89],{"name":78,"url":76},[21,91,22,92,93],"农业机器人","模仿学习","草莓采摘",{"code":95,"slug":96,"name":97,"path":98,"parent_code":99},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables","B",[],{"status":102,"locked":14,"version":27,"method":28,"evidence":103},"classified",{"candidates":104,"agricultural_evidence":34},[105,111,116,120],{"code":95,"score":106,"supported":34,"title":107,"abstract":109,"tags":110},8,[108],"草莓",[],[],{"code":112,"score":86,"supported":34,"title":113,"abstract":114,"tags":115},"F6",[],[91],[91],{"code":32,"score":43,"supported":14,"title":117,"abstract":118,"tags":119},[],[],[21],{"code":42,"score":43,"supported":14,"title":121,"abstract":122,"tags":123},[],[],[22],[125,126],"草莓果柄 定点采摘 机械臂","Jetson AGX Orin 采摘 模仿学习","草莓果柄定点采摘机械臂-5533",{"doi":9,"openalex_id":9,"authors":129,"venue":9,"cited_by_count":15,"oa_url":9,"card":130,"direction":134,"ingested_from":136},[],{"tldr":131,"method":132,"finding":133,"direction":134,"opportunity":135},"用低成本遥操作机械臂采集演示，模仿学习实现草莓果柄定点采摘。","ResNet-50骨干的动作分块Transformer策略，Jetson边缘部署","130条演示训练后损失降至0.091，策略可15赫兹自主采摘。","农业人工智能与决策模型","可探索小样本跨品种泛化与失败恢复，降低演示采集成本。","agent","2026-10-22T00:07:09.891422Z",{"id":139,"title":140,"url":141,"summary":142,"summary_zh":143,"content":9,"source_name":144,"source_url":141,"published_at":145,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":146,"score_detail":147,"sources":151,"tags":153,"primary_topic":9,"secondary_topics":157,"classification":158,"search_phrases":193,"slug":196,"view_count":15,"doi":197,"paper":198,"created_at":221},5402,"AI‐Enabled RGB Light‐to‐Frequency Sensor for Nondestructive Salinity Stress Monitoring in Medicinal Plants","https:\u002F\u002Fdoi.org\u002F10.1002\u002Faisy.70561","Salinity stress constrains medicinal plant productivity, particularly in coastal and salt affected agroecosystems where rapid, deployable diagnostics are essential. Here, we develop an AI‐enabled RGB light‐to‐frequency proximal sensing platform integrated with Tiny Machine Learning (TinyML) for nondestructive, real‐time salinity discrimination. Plants are exposed to three salinity levels (0, 100, and 200 mM NaCl). RGB‐derived and CIELAB chromatic descriptors are engineered, normalized, and partitioned (75:25) into training and test datasets. ReliefF‐based feature ranking and systematic subset optimization are combined with supervised learning using fivefold cross‐validation, and interpretability is achieved through SHAP analysis. Biomass reductions under saline treatments confirm physiological perturbations underlying spectral variability. Subspace KNN achieved the highest predictive performance (96.28% validation; 95.55% independent test accuracy). Framing model selection as a multiobjective optimization problem identified an optimal 10‐feature subset. Under this constraint, a Wide Neural Network attained 0.904 accuracy with an ultracompact 0.017 MB model size and >246 000 observations s −1 inference speed, demonstrating full TinyML compatibility. SHAP attribution consistently highlighted green intensity, blue channel metrics, and chromatic ratios as dominant predictors. This work establishes a lightweight, interpretable Edge‐AI framework for scalable, real‐time salinity diagnostics in precision agriculture.","盐胁迫制约药用植物的生产力，尤其是在沿海及受盐影响的农业生态系统中，亟需可快速部署的诊断手段。本研究开发了一种基于人工智能的RGB光-频率近端传感平台，并集成微型机器学习（TinyML），用于无损、实时的盐胁迫判别。植物暴露于三个盐浓度水平（0、100和200 mM NaCl）。研究构建了RGB衍生描述符与CIELAB色度描述符，经归一化后按75:25划分为训练集与测试集。将基于ReliefF的特征排序与系统性特征子集优化相结合，采用五折交叉验证进行监督学习，并通过SHAP分析实现模型可解释性。盐处理下的生物量下降证实了光谱变异性背后的生理扰动。子空间KNN取得了最高的预测性能（验证集准确率96.28%；独立测试集准确率95.55%）。将模型选择构建为多目标优化问题，确定了最优的10特征子集。在该约束下，宽神经网络（Wide Neural Network）达到0.904的准确率，模型体积仅0.017 MB，推理速度超过246 000次观测\u002F秒，展现出完全的TinyML兼容性。SHAP归因一致表明，绿色强度、蓝色通道指标及色度比是主要预测因子。本研究为精准农业中可扩展的实时盐胁迫诊断建立了一个轻量、可解释的边缘人工智能框架。","Advanced Intelligent Systems","2026-10-19T00:00:00Z",81,{"impact":84,"substance":148,"depth":84,"authority":149,"freshness":85,"relevant":43,"comment":150},22,14,"该研究提出一种轻量、可解释的边缘AI盐胁迫诊断框架，方法新颖、数据扎实，对精准农业具有参考价值，但属于实验室阶段成果，产业影响有限。",[152],{"name":144,"url":141},[154,21,22,155,156],"智慧农业","盐胁迫监测","药用植物",[],{"status":26,"locked":14,"version":27,"method":28,"evidence":159},{"candidates":160,"agricultural_evidence":34,"reason":192},[161,168,175,181,188],{"code":162,"score":163,"supported":34,"title":164,"abstract":165,"tags":167},"B6",7,[],[156,166],"medicinal plant",[156],{"code":169,"score":163,"supported":34,"title":170,"abstract":171,"tags":174},"I1",[],[172,173],"精准农业","precision agriculture",[154],{"code":32,"score":163,"supported":34,"title":176,"abstract":177,"tags":180},[],[178,179],"机器学习","machine learning",[21],{"code":182,"score":183,"supported":34,"title":184,"abstract":185,"tags":187},"H6",3,[],[186],"农业生态",[],{"code":42,"score":43,"supported":14,"title":189,"abstract":190,"tags":191},[],[],[22],"多个产业证据相当，需要确认主归属",[194,195],"RGB 光频传感器 盐胁迫","TinyML 药用植物 无损监测","RGB光频传感器盐胁迫-5402","10.1002\u002Faisy.70561",{"doi":197,"openalex_id":199,"authors":200,"venue":144,"cited_by_count":15,"oa_url":141,"card":216,"direction":134,"ingested_from":68},"W7222914402",[201,204,207,210,213],{"name":202,"orcid":203},"Ahmad Ali","https:\u002F\u002Forcid.org\u002F0000-0001-5530-7374",{"name":205,"orcid":206},"Vinie Lee Silva-Alvarado","https:\u002F\u002Forcid.org\u002F0009-0000-5857-3248",{"name":208,"orcid":209},"Arman Heydari","https:\u002F\u002Forcid.org\u002F0009-0002-4705-0729",{"name":211,"orcid":212},"Sandra Sendra","https:\u002F\u002Forcid.org\u002F0000-0001-9556-9088",{"name":214,"orcid":215},"Jaime Lloret","https:\u002F\u002Forcid.org\u002F0000-0002-0862-0533",{"tldr":217,"method":218,"finding":219,"direction":66,"opportunity":220},"开发AI驱动的RGB光频传感器，实现药用植物盐胁迫无损实时监测。","RGB与CIELAB色度特征、ReliefF特征选择、SHAP解释及TinyML","Subspace KNN验证准确率96.28%，10特征Wide NN模型仅0.017MB且推理超2","可探索多作物多胁迫通用色度特征库，及田间光照变化下的模型鲁棒性与长期部署验证。","2026-10-20T23:30:35.809539Z",{"id":223,"title":224,"url":225,"summary":226,"summary_zh":227,"content":9,"source_name":228,"source_url":225,"published_at":229,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":230,"score_detail":231,"sources":235,"tags":237,"primary_topic":240,"secondary_topics":245,"classification":246,"search_phrases":270,"slug":273,"view_count":15,"doi":274,"paper":275,"created_at":295},5401,"A HIERARCHICAL DISTRIBUTED EDGE AI FRAMEWORK FOR REAL-TIME APPLE LEAF DISEASE CLASSIFICATION","https:\u002F\u002Fdoi.org\u002F10.22190\u002Ffuacr260506002g","The application of artificial intelligence (AI) in agriculture has transformed crop management, particularly in the detection of plant diseases. Apple trees, a key agricultural commodity, are vulnerable to various diseases, which can severely impact crop yields. Traditional disease detection methods, often reliant on manual inspections, are time-consuming and prone to error. In response, this study proposes a novel two-stage edge AI framework for apple leaf disease classification that integrates RGB imagery with distributed edge computing. The first stage performs binary classification to distinguish between healthy and diseased leaves, achieving an F1 score of 95.50%. This optimization is achieved by utilizing lightweight models on edge devices, thereby enhancing computational efficiency. The second stage employs a consensus-based multiclass classification approach, achieving a 96.44% F1 score in identifying specific diseases. This dual-stage system, supported by an edge AI architecture, ensures real-time disease detection while minimizing resource consumption and enhancing scalability in large-scale agricultural environments. The full system achieved a 95.50% accuracy and F1 score on the special dataset. The results demonstrate the system’s efficacy in controlled laboratory conditions, offering a promising solution for precise, real-time agricultural monitoring. This study underscores the potential of distributed edge AI systems for scalable and efficient disease detection in precision agriculture.","人工智能（AI）在农业中的应用已深刻改变了作物管理方式，尤其是在植物病害检测方面。苹果树作为重要的农业经济作物，易受多种病害侵袭，严重影响作物产量。传统的病害检测方法通常依赖人工检查，耗时且易出错。为此，本研究提出了一种新颖的两阶段边缘AI框架，用于苹果叶片病害分类，该框架将RGB图像与分布式边缘计算相结合。第一阶段执行二分类以区分健康叶片与患病叶片，F1分数达到95.50%。该优化通过在边缘设备上采用轻量级模型实现，从而提升了计算效率。第二阶段采用基于共识的多类分类方法，在识别具体病害方面F1分数达到96.44%。该双阶段系统在边缘AI架构的支持下，实现了实时病害检测，同时最大限度地降低了资源消耗，并增强了在大规模农业环境中的可扩展性。完整系统在专用数据集上达到了95.50%的准确率和F1分数。结果表明，该系统在受控实验室条件下具有有效性，为精准、实时的农业监测提供了一种有前景的解决方案。本研究凸显了分布式边缘AI系统在精准农业中实现可扩展且高效病害检测的潜力。","Facta Universitatis Series Automatic Control and Robotics","2026-10-18T00:00:00Z",76,{"impact":84,"substance":83,"depth":232,"authority":233,"freshness":106,"relevant":43,"comment":234},17,12,"该研究提出两级边缘AI框架，在苹果叶片病害识别上取得95%以上F1值，方法新颖、数据可靠，对精准农业实时监测有参考价值，但尚处实验室验证阶段，产业影响有限。",[236],{"name":228,"url":225},[154,21,22,238,239],"作物病害识别","苹果病害",{"code":241,"slug":242,"name":243,"path":244,"parent_code":99},"B2","fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit",[],{"status":102,"locked":14,"version":27,"method":28,"evidence":247},{"candidates":248,"agricultural_evidence":34},[249,255,259,264],{"code":241,"score":84,"supported":34,"title":250,"abstract":252,"tags":254},[251],"apple",[243,253,251],"苹果",[239],{"code":169,"score":163,"supported":34,"title":256,"abstract":257,"tags":258},[],[172,173],[154],{"code":42,"score":163,"supported":34,"title":260,"abstract":261,"tags":263},[],[22,262],"edge computing",[22],{"code":32,"score":265,"supported":34,"title":266,"abstract":267,"tags":269},5,[],[268],"artificial intelligence",[238,21],[271,272],"苹果叶片病害 边缘AI","分布式边缘计算 病害分类","苹果叶片病害边缘AI-5401","10.22190\u002Ffuacr260506002g",{"doi":274,"openalex_id":276,"authors":277,"venue":228,"cited_by_count":15,"oa_url":225,"card":290,"direction":134,"ingested_from":68},"W7222964752",[278,281,284,287],{"name":279,"orcid":280},"Milan Grujev","https:\u002F\u002Forcid.org\u002F0009-0002-3211-6844",{"name":282,"orcid":283},"Dragan H. Stojanovic","https:\u002F\u002Forcid.org\u002F0000-0003-4893-3814",{"name":285,"orcid":286},"Miloš Ilić","https:\u002F\u002Forcid.org\u002F0000-0002-9696-2874",{"name":288,"orcid":289},"Aleksandar Milosavljević","https:\u002F\u002Forcid.org\u002F0000-0003-3057-7727",{"tldr":291,"method":292,"finding":293,"direction":66,"opportunity":294},"提出两级边缘AI框架，用RGB图像实时分类苹果叶片病害。","分布式边缘计算+轻量模型，先二分类再共识多分类。","两级系统准确率与F1均达95.50%，多分类F1为96.44%。","仅在受控实验室验证，可延伸至田间复杂光照与多作物泛化及边缘能耗优化。","2026-10-20T23:30:33.830748Z",{"id":297,"title":298,"url":299,"summary":300,"summary_zh":301,"content":9,"source_name":302,"source_url":299,"published_at":229,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":303,"score_detail":304,"sources":306,"tags":308,"primary_topic":311,"secondary_topics":312,"classification":318,"search_phrases":338,"slug":341,"view_count":15,"doi":342,"paper":343,"created_at":357},5387,"LIGHTWEIGHT FEATURE FUSED ATTENTION DRIVEN MULTI LEVEL HYBRID MOBILENET V2 FRAMEWORK FOR EARLY TOMATO LEAF DISEASE DETECTION","https:\u002F\u002Fdoi.org\u002F10.62441\u002Factainnovations.vi.1918","It is crucial to find and treat tomato crop diseases early and accurately to improve crop efficiency and make sure there is no scarcity of crop. However, computerized analysis is provocative due to lighting differences, compound backgrounds, and graphic likenesses among infection regions. Research work proposed a lightweight Attention Driven Multi Level Hybrid MobileNet-V2 algorithm that fuses the handcrafted texture structures with Deep Learning (DL) features, then image preprocessing is incorporated using Gaussian, Bilateral, and Guided filters to minimize noise by preserving the disease lesions. LAB color model along with Contrast Limited Adaptive Histogram Equalization (CLAHE) are used to improve the contrast. Otsu thresholding is involved to effectively separate the leaf segment from diseased parts. Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) descriptors are used to obtain fine-grained features, while high level features are gained by pretrained MobileNet-V2 model. Features fusion is fed to the fully connected layers of deep learning model for training and validation purpose. Results have revealed that the suggested model attains better classification performance as compared to existing models, making it an effective explanation for real time tomato leaf disease detection for smart agriculture applications.","早期准确发现并治疗番茄作物病害，对于提高作物效率、确保作物不短缺至关重要。然而，由于光照差异、复杂背景以及感染区域之间的图形相似性，计算机化分析具有挑战性。研究提出了一种轻量级注意力驱动多层混合MobileNet-V2算法，该算法将手工纹理结构与深度学习（DL）特征相融合，然后结合高斯、双边和引导滤波器进行图像预处理，以在保留病害病变的同时最小化噪声。采用LAB颜色模型和对比度受限自适应直方图均衡化（CLAHE）来提高对比度。引入Otsu阈值分割以有效分离叶片片段与病害部分。使用局部二值模式（LBP）和灰度共生矩阵（GLCM）描述符获取细粒度特征，同时通过预训练的MobileNet-V2模型获取高层特征。特征融合后输入深度学习模型的全连接层进行训练和验证。结果表明，与现有模型相比，所提出的模型取得了更好的分类性能，使其成为智慧农业应用中实时番茄叶片病害检测的有效解决方案。","Acta Innovations",64,{"impact":233,"substance":84,"depth":82,"authority":13,"freshness":106,"relevant":43,"comment":305},"提出融合手工纹理特征与MobileNet-V2的轻量番茄叶病检测方法，有一定方法新颖性，但属单篇论文、应用范围有限，未达每日精选门槛。",[307],{"name":302,"url":299},[154,21,309,310,23],"深度学习","番茄病害",{"code":95,"slug":96,"name":97,"path":98,"parent_code":99},[313],{"code":32,"slug":314,"name":315,"path":316,"parent_code":317},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":102,"locked":14,"version":27,"method":28,"evidence":319},{"candidates":320,"agricultural_evidence":34},[321,328,333],{"code":95,"score":322,"supported":34,"title":323,"abstract":325,"tags":327},15,[324],"tomato",[326,324],"番茄",[310],{"code":32,"score":85,"supported":34,"title":329,"abstract":330,"tags":332},[],[309,331],"deep learning",[21,23,309],{"code":169,"score":163,"supported":34,"title":334,"abstract":335,"tags":337},[],[154,336],"smart agriculture",[154],[339,340],"MobileNet-V2 番茄叶片病害","番茄叶病 深度学习 检测","MobileNet-V2番茄叶片病害-5387","10.62441\u002Factainnovations.vi.1918",{"doi":342,"openalex_id":344,"authors":345,"venue":302,"cited_by_count":15,"oa_url":351,"card":352,"direction":66,"ingested_from":68},"W7222977034",[346,348],{"name":347,"orcid":9},"Mrs. Vijetalaxmi A Hegde",{"name":349,"orcid":350},"Abhay Ramakant Kasetwar","https:\u002F\u002Forcid.org\u002F0000-0002-5800-8764","https:\u002F\u002Factainnovations.com\u002Findex.php\u002Fpub\u002Farticle\u002Fdownload\u002F1918\u002F981",{"tldr":353,"method":354,"finding":355,"direction":134,"opportunity":356},"提出轻量注意力多级混合MobileNet-V2框架，融合手工纹理与深度学习特征实现番茄叶片病害早期检","高斯\u002F双边\u002F引导滤波去噪，LAB+CLAHE增强，Otsu分割，LBP与GLCM","所提模型分类性能优于现有模型，可有效用于智慧农业实时番茄叶片病害检测。","可探索多作物跨域泛化、田间复杂光照下轻量模型部署与可解释性，及病害早期分级检测。","2026-10-20T23:30:10.424511Z",{"id":359,"title":360,"url":361,"summary":362,"summary_zh":363,"content":9,"source_name":364,"source_url":361,"published_at":365,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":366,"score_detail":367,"sources":370,"tags":372,"primary_topic":375,"secondary_topics":376,"classification":377,"search_phrases":403,"slug":406,"view_count":43,"doi":407,"paper":408,"created_at":427},5248,"An entropy-guided dynamic lightweight network for identification of wild edible vegetables of BTR","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02528-x","Wild edible vegetables of Bodoland Territorial Region (BTR) are the species of plants that grow naturally and are commonly consumed as food by the indigenous communities. These plants are important sources of nutritional and therapeutic properties and are essential for food security and socio-economic development of rural communities. Therefore, sustainable conservation and utilization of these unique plant resources are of great significance. Developing an efficient and accurate automatic identification system is an essential conservation strategy for them. Current plant identification methods are mainly based on classical machine learning (ML) methods and deep convolutional neural network (CNN) architectures for recognition and classification. These methods are effective for images taken under controlled or pristine conditions, but the performance is often far from ideal in the complex and natural environments of forest and grassland, where wild edible plants are found. This paper proposes an entropy-guided dynamic lightweight deep learning architecture for automatic detection of wild edible vegetables in BTR. The proposed model employs adaptive computation with Dynamic Kernel Selection (DKS), Region-wise Adaptive Feature Routing (RAFR), Ultra-Light Context Encoding (ULCE), and Entropy-Guided Global Pooling (EGGP) to capture multi-scale plant characteristics efficiently and reduce redundant computations. The novelty of our proposed framework lies in its coordinated integration of these modules that is designed to reduce computational overhead without compromising with performance. Experimental results show that the proposed model achieves excellent classification accuracy of 90.69% with significantly reduced parameter complexity and floating-point operations (FLOPs), which is well-suited for real-time precision agriculture and low-resource edge deployment applications.","博多兰领土区域（BTR）的野生食用蔬菜是自然生长并被土著社区普遍作为食物食用的植物物种。这些植物是营养和治疗特性的重要来源，对农村社区的粮食安全和社会经济发展至关重要。因此，对这些独特植物资源的可持续保护和利用具有重要意义。开发高效准确的自动识别系统是保护这些植物资源的重要策略。当前的植物识别方法主要基于经典机器学习（ML）方法和深度卷积神经网络（CNN）架构进行识别和分类。这些方法对于在受控或原始条件下拍摄的图像是有效的，但在野生食用植物所处的森林和草地复杂自然环境中，其性能往往远不理想。本文提出了一种熵引导的动态轻量级深度学习架构，用于自动检测BTR中的野生食用蔬菜。所提出的模型采用动态核选择（DKS）、区域自适应特征路由（RAFR）、超轻量上下文编码（ULCE）和熵引导全局池化（EGGP）的自适应计算，以高效捕捉多尺度植物特征并减少冗余计算。我们所提出框架的新颖性在于这些模块的协同集成，旨在在不影响性能的情况下降低计算开销。实验结果表明，所提出的模型达到了90.69%的优异分类准确率，同时显著降低了参数复杂度和浮点运算量（FLOPs），非常适合实时精准农业和低资源边缘部署应用。","Discover Artificial Intelligence","2026-10-17T00:00:00Z",70,{"impact":233,"substance":368,"depth":232,"authority":233,"freshness":85,"relevant":43,"comment":369},20,"提出熵引导动态轻量网络用于野生食用蔬菜识别，准确率90.69%且参数与算力显著降低，方法有新意、结论可靠，但属细分应用研究，公共影响有限。",[371],{"name":364,"url":361},[154,21,22,373,374],"野生蔬菜","植物识别",{"code":32,"slug":314,"name":315,"path":316,"parent_code":317},[],{"status":102,"locked":14,"version":27,"method":28,"evidence":378},{"candidates":379,"agricultural_evidence":34},[380,384,388,395,399],{"code":32,"score":13,"supported":34,"title":381,"abstract":382,"tags":383},[],[309,178,179,331],[21],{"code":169,"score":163,"supported":34,"title":385,"abstract":386,"tags":387},[],[172,173],[154],{"code":389,"score":71,"supported":34,"title":390,"abstract":391,"tags":394},"A8",[],[392,393],"粮食安全","food security",[],{"code":95,"score":183,"supported":34,"title":396,"abstract":397,"tags":398},[],[97],[],{"code":42,"score":43,"supported":14,"title":400,"abstract":401,"tags":402},[],[],[22],[404,405],"BTR 野生食用蔬菜 识别","熵引导 轻量网络 植物识别","BTR野生食用蔬菜识别-5248","10.1007\u002Fs44163-026-02528-x",{"doi":407,"openalex_id":409,"authors":410,"venue":364,"cited_by_count":15,"oa_url":361,"card":422,"direction":134,"ingested_from":68},"W7222842686",[411,413,416,419],{"name":412,"orcid":9},"Tania Sarkar",{"name":414,"orcid":415},"Pranav Kumar Singh","https:\u002F\u002Forcid.org\u002F0000-0001-8987-0229",{"name":417,"orcid":418},"Parag Bhuyan","https:\u002F\u002Forcid.org\u002F0009-0006-4430-1851",{"name":420,"orcid":421},"Anshuman Kalla","https:\u002F\u002Forcid.org\u002F0000-0002-3985-4578",{"tldr":423,"method":424,"finding":425,"direction":134,"opportunity":426},"提出熵引导动态轻量网络，用于识别BTR野生食用蔬菜。","动态核选择、区域自适应特征路由、超轻上下文编码与熵引导池化。","分类准确率达90.69%，参数量和FLOPs显著降低，适合边缘部署。","可探索复杂自然场景下轻量模型在低资源边缘设备的实时植物识别与保护应用。","2026-10-18T23:30:58.805321Z",{"id":429,"title":430,"url":431,"summary":432,"summary_zh":433,"content":9,"source_name":434,"source_url":431,"published_at":365,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":435,"sources":437,"tags":439,"primary_topic":9,"secondary_topics":440,"classification":441,"search_phrases":448,"slug":450,"view_count":15,"doi":451,"paper":452,"created_at":464},5242,"Enhanced Convolutional Neural Network Architecture with Attention-Guided Feature Refinement for Robust Image Classification","https:\u002F\u002Fdoi.org\u002F10.25258\u002Fijddt.16.88s.43","Image classification remains a foundational problem in computer vision, underpinning applications that range from medical diagnostics to autonomous navigation. Although deep convolutional neural networks (CNNs) have substantially advanced classification accuracy over the last decade, standard architectures such as VGG, ResNet, and plain CNNs often struggle to jointly optimize accuracy, parameter efficiency, and robustness to intra-class variation. This paper proposes an AttentionGuided Refinement Network (AGR-Net), a CNN architecture that integrates a lightweight channel-and-spatial attention module with residual feature refinement blocks to improve discriminative feature learning while maintaining a compact parameter footprint. The proposed model is evaluated on the CIFAR-10 benchmark dataset using a rigorous experimental protocol that includes data augmentation, learning-rate scheduling, and five-fold cross-validation. Comparative experiments against baseline architectures, including a plain CNN, VGG-16, ResNet-18, and MobileNetV2, demonstrate that AGR-Net achieves a top-1 accuracy of 94.6%, outperforming the strongest baseline by 1.8 percentage points while using 27% fewer parameters than ResNet-18. Ablation studies confirm that the attention module and refinement blocks each contribute measurably to the accuracy gain. The results indicate that attention-guided refinement offers a favorable trade-off between accuracy, efficiency, and robustness, making the proposed architecture suitable for deployment in resource-constrained and real-time image classification systems. The paper further discusses the advantages, practical applications, limitations, and future research directions of the proposed approach.","图像分类仍然是计算机视觉中的一个基础性问题，支撑着从医学诊断到自主导航等诸多应用。尽管深度卷积神经网络（CNN）在过去十年中大幅提升了分类准确率，但VGG、ResNet和普通CNN等标准架构往往难以在准确率、参数效率和对类内变化的鲁棒性之间实现联合优化。本文提出了一种注意力引导细化网络（AttentionGuided Refinement Network，AGR-Net），这是一种CNN架构，它将轻量级的通道与空间注意力模块与残差特征细化模块相结合，以改进判别性特征学习，同时保持紧凑的参数规模。所提出的模型在CIFAR-10基准数据集上使用严格的实验方案进行评估，该方案包括数据增强、学习率调度和五折交叉验证。与基线架构（包括普通CNN、VGG-16、ResNet-18和MobileNetV2）的对比实验表明，AGR-Net实现了94.6%的top-1准确率，比最强基线高出1.8个百分点，同时使用的参数比ResNet-18少27%。消融研究证实，注意力模块和细化模块各自对准确率提升都有可测量的贡献。结果表明，注意力引导细化在准确率、效率和鲁棒性之间提供了有利的权衡，使所提出的架构适合部署在资源受限和实时的图像分类系统中。本文进一步讨论了所提出方法的优势、实际应用、局限性和未来研究方向。","International Journal of Drug Delivery Technology",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":436},"该文为通用计算机视觉论文，未涉及农业场景，与三农及农业信息化无关，不宜入选。",[438],{"name":434,"url":431},[21,309,23],[],{"status":26,"locked":14,"version":27,"method":28,"evidence":442},{"candidates":443,"agricultural_evidence":14,"reason":47},[444],{"code":32,"score":85,"supported":34,"title":445,"abstract":446,"tags":447},[],[38,39],[21,23,309],[449,50],"农业人工智能 图像识别 深度学习","农业人工智能图像识别深度学习-5242","10.25258\u002Fijddt.16.88s.43",{"doi":451,"openalex_id":453,"authors":454,"venue":434,"cited_by_count":15,"oa_url":431,"card":459,"direction":66,"ingested_from":68},"W7222796569",[455,457],{"name":456,"orcid":9},"KRISHNA SANDEEP JONNALA",{"name":458,"orcid":9},"K VENKAT RAJU",{"tldr":460,"method":461,"finding":462,"direction":134,"opportunity":463},"提出注意力引导细化网络AGR-Net，在CIFAR-10上兼顾精度与参数效率。","轻量通道-空间注意力模块+残差特征细化块，五折交叉验证对比基线。","AGR-Net达94.6%准确率，超最强基线1.8个百分点且参数比ResNet-18少27%。","可将该轻量注意力架构迁移至作物病害识别等农业图像任务，验证田间复杂场景鲁棒性。","2026-10-18T23:30:13.674128Z"]