[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2940":3,"related-2940":113},{"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":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":44,"search_phrases":92,"slug":95,"view_count":96,"doi":97,"paper":98,"created_at":112},2940,"A Hybrid CNN–Transformer–LSTM Deep Learning Framework for Automated Cotton Disease Detection","https:\u002F\u002Fdoi.org\u002F10.53365\u002Fnrfhh.1816","Cotton production is frequently affected by leaf diseases that can reduce plant productivity, deteriorate crop quality, and cause considerable financial losses for farmers. Consequently, rapid and reliable disease identification is an important requirement for precision agriculture and effective crop protection. Conventional image-based approaches predominantly employ CNN architectures for visual classification; however, such methods may have limited capability in learning long-range spatial dependencies and modeling changes associated with disease development over time. To overcome these limitations, this research introduces an integrated hybrid deep learning framework that combines Convolutional Neural Networks (CNNs), Transformer-based self-attention, and Long Short-Term Memory (LSTM) networks for intelligent cotton leaf disease recognition. The proposed framework utilizes a pre-trained EfficientNet network to extract discriminative spatial characteristics from cotton leaf images. The extracted representations are subsequently processed through a multi-head self-attention Transformer module, which enables the network to identify relationships between distant and relevant regions of the leaf. An LSTM component is then incorporated to learn sequential dependencies and provide a foundation for analyzing disease evolution and progression. To improve model reliability and generalization, the network is trained using an appropriately organized cotton leaf image dataset together with image augmentation, transfer learning, and fine-tuning techniques. The experimental evaluation indicates that the proposed CNN–Transformer–LSTM architecture provides improved disease classification performance and more effective feature learning compared with conventional CNN-based models. Performance assessment using classification metrics, confusion matrices, and ROC curves demonstrates strong discrimination among the considered cotton disease categories. The combined learning of local visual characteristics, global contextual information, and sequential dependencies provides a comprehensive framework for automated cotton disease assessment. The proposed approach can serve as a scalable artificial intelligence solution for precision agriculture applications. Its architecture also provides opportunities for future integration with IoT-based crop monitoring systems, environmental data acquisition, and disease progression prediction, supporting early warning systems and intelligent crop health management in smart farming environments.","棉花生产常受叶片病害影响，这些病害会降低植株生产力、恶化作物品质，并给农民造成相当大的经济损失。因此，快速可靠的病害识别是精准农业和有效作物保护的重要需求。传统的基于图像的方法主要采用CNN架构进行视觉分类；然而，此类方法在学习长程空间依赖关系以及建模与病害随时间发展相关的变化方面能力有限。为克服这些局限，本研究提出了一种集成混合深度学习框架，将卷积神经网络（CNN）、基于Transformer的自注意力机制和长短期记忆（LSTM）网络相结合，用于智能棉花叶片病害识别。所提出的框架利用预训练的EfficientNet网络从棉花叶片图像中提取具有判别力的空间特征。提取到的表示随后通过多头自注意力Transformer模块进行处理，使网络能够识别叶片中相距较远且相关区域之间的关系。随后引入LSTM组件以学习序列依赖关系，并为分析病害演变和进展提供基础。为提高模型可靠性和泛化能力，网络使用组织良好的棉花叶片图像数据集进行训练，并结合图像增强、迁移学习和微调技术。实验评估表明，与传统的基于CNN的模型相比，所提出的CNN–Transformer–LSTM架构提供了更好的病害分类性能和更有效的特征学习。使用分类指标、混淆矩阵和ROC曲线进行的性能评估表明，该方法在所考虑的棉花病害类别之间具有较强的判别能力。局部视觉特征、全局上下文信息和序列依赖关系的联合学习为自动化棉花病害评估提供了一个全面的框架。所提出的方法可作为精准农业应用中可扩展的人工智能解决方案。其架构也为未来与基于物联网的作物监测系统、环境数据采集和病害进展预测的集成提供了机会，从而支持预警系统和智能",null,"Natural Resources for Human Health","2026-09-16T00:00:00Z","论文",10,false,69,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":22},12,20,17,8,1,"提出CNN-Transformer-LSTM混合框架用于棉花叶病识别，方法有创新但属实验室验证阶段，产业影响有限。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","深度学习","精准农业","棉花病害识别",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[38],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":45,"locked":14,"version":46,"method":47,"evidence":48},"classified","2026-10-22.2","rules",{"candidates":49,"agricultural_evidence":52},[50,58,65,71,79,86],{"code":39,"score":51,"supported":52,"title":53,"abstract":55,"tags":57},19,true,[54],"deep learning",[28,54,56],"artificial intelligence",[27,28],{"code":32,"score":59,"supported":52,"title":60,"abstract":62,"tags":64},14,[61],"cotton",[63,61],"棉花",[],{"code":66,"score":20,"supported":52,"title":67,"abstract":68,"tags":70},"I1",[],[29,69],"precision agriculture",[26,29],{"code":72,"score":73,"supported":52,"title":74,"abstract":75,"tags":78},"I5",6,[],[76,77],"物联网","IoT",[],{"code":80,"score":81,"supported":52,"title":82,"abstract":83,"tags":85},"A4",3,[],[84],"花生",[],{"code":87,"score":81,"supported":52,"title":88,"abstract":89,"tags":91},"F4",[],[90],"crop protection",[],[93,94],"棉花叶病 CNN Transformer LSTM","EfficientNet 棉花病害检测","棉花叶病CNNTransformerLSTM-2940",5,"10.53365\u002Fnrfhh.1816",{"doi":97,"openalex_id":99,"authors":100,"venue":10,"cited_by_count":103,"oa_url":9,"card":104,"direction":110,"ingested_from":111},"W7213544712",[101],{"name":102,"orcid":9},"Prajakta Sunil Gupta",0,{"tldr":105,"method":106,"finding":107,"direction":108,"opportunity":109},"提出CNN-Transformer-LSTM混合框架，实现棉花叶片病害自动识别。","EfficientNet提取特征，Transformer自注意力与LSTM建模，","混合模型分类性能优于传统CNN，能同时学习局部特征、全局上下文和序列依赖。","农业人工智能与决策模型","可融合IoT环境数据与时间序列，开展病害进展预测和早期预警系统研究。","智慧农业 \u002F 农业物联网","openalex","2026-09-19T23:30:19.558449Z",{"total":73,"page":21,"page_size":73,"items":114},[115,197,273,336,388,440],{"id":116,"title":117,"url":118,"summary":119,"summary_zh":120,"content":9,"source_name":121,"source_url":118,"published_at":122,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":123,"score_detail":124,"sources":130,"tags":132,"primary_topic":135,"secondary_topics":136,"classification":138,"search_phrases":157,"slug":160,"view_count":21,"doi":161,"paper":162,"created_at":196},5150,"MSFANet: a multi-scale feature aggregation network for tobacco pest recognition","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1871587","Introduction Tobacco is a globally significant economic crop, and pest infestation represents one of the primary factors contributing to declines in both yield and quality. However, tobacco pests encompass a wide variety of species that differ substantially in morphology and scale. Field-captured images are frequently subject to complex background interference and class imbalance, presenting considerable challenges for automated recognition. Existing deep learning approaches remain limited in their capacity for multi-scale feature extraction and discriminative feature representation under such conditions. To address these challenges, this study proposes a Multi-Scale Feature Aggregation Network (MSFANet) for tobacco pest recognition, formulated here as a single-label image classification task. Methods To address scale variation and complex field backgrounds, two task-oriented modules were developed. The Multi-Scale Pyramid Pooling (MSPP) module aggregates contextual information across multiple receptive fields through parallel feature extraction and adaptive cross-branch fusion. The Hierarchical Feature Fusion Block (HFFB) subsequently refines the feature representations through spatial gating and partial convolution. An auxiliary classification head and Focal Loss were incorporated during training to provide intermediate supervision and address class imbalance, respectively. Results The effectiveness of MSFANet was evaluated on a dataset comprising 13 tobacco-associated pest and natural-enemy categories. The model achieved accuracy, precision, recall, and F1-score of 97.63%, 97.89%, 97.59%, and 97.71%, respectively, with a parameter count of 1.21 M. Among the models trained from random initialization, MSFANet shared the highest accuracy while using the fewest parameters. Discussion By combining multi-scale contextual aggregation with hierarchical feature refinement, MSFANet achieved accurate classification across the 13 tobacco-associated categories under complex field conditions. The proposed method provides effective technical support for automated and precision-oriented pest monitoring in tobacco fields, offering practical value for advancing precise pest management and the broader development of smart agriculture.","引言 烟草是全球重要的经济作物，虫害侵染是导致其产量和品质下降的主要因素之一。然而，烟草害虫种类繁多，形态和尺度差异显著。田间采集的图像常受到复杂背景干扰和类别不平衡的影响，给自动化识别带来了相当大的挑战。现有深度学习方法在此类条件下进行多尺度特征提取和判别性特征表示的能力仍然有限。为应对这些挑战，本研究提出了一种用于烟草害虫识别的多尺度特征聚合网络（Multi-Scale Feature Aggregation Network，MSFANet），并将其形式化为单标签图像分类任务。方法 针对尺度变化和复杂田间背景，开发了两个面向任务的模块。多尺度金字塔池化（Multi-Scale Pyramid Pooling，MSPP）模块通过并行特征提取和自适应跨分支融合，聚合多个感受野的上下文信息。分层特征融合模块（Hierarchical Feature Fusion Block，HFFB）随后通过空间门控和部分卷积对特征表示进行细化。训练过程中引入辅助分类头和Focal Loss，分别提供中间监督和处理类别不平衡。结果 在包含13个烟草相关害虫及天敌类别的数据集上评估了MSFANet的有效性。模型的准确率、精确率、召回率和F1分数分别为97.63%、97.89%、97.59%和97.71%，参数量为1.21 M。在随机初始化训练的模型中，MSFANet以最少的参数量取得了最高准确率。讨论 通过结合多尺度上下文聚合与分层特征细化，MSFANet在复杂田间条件下实现了对13个烟草相关类别的准确分类。所提方法为烟草田间害虫的自动化和精准监测提供了有效的技术支持，对推进精准害虫管理和智慧农业的更广泛发展具有实用价值。","Frontiers in Plant Science","2026-10-16T00:00:00Z",76,{"impact":125,"substance":126,"depth":19,"authority":127,"freshness":128,"relevant":21,"comment":129},16,21,13,9,"该论文提出轻量化多尺度特征聚合网络，在烟草害虫识别上取得97%以上精度且参数量仅1.21M，方法新颖、数据扎实，对智慧农业虫情监测有实用价值，值得进入每日精选。",[131],{"name":121,"url":118},[26,27,28,133,134],"病虫害识别","烟草",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[137],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":139},{"candidates":140,"agricultural_evidence":52},[141,147,151],{"code":32,"score":142,"supported":52,"title":143,"abstract":145,"tags":146},15,[144],"tobacco",[134,144],[134],{"code":39,"score":128,"supported":52,"title":148,"abstract":149,"tags":150},[],[28,54],[27,28,133],{"code":66,"score":152,"supported":52,"title":153,"abstract":154,"tags":156},7,[],[26,155],"smart agriculture",[26],[158,159],"MSFANet 烟草害虫识别","多尺度特征聚合 害虫识别","MSFANet烟草害虫识别-5150","10.3389\u002Ffpls.2026.1871587",{"doi":161,"openalex_id":163,"authors":164,"venue":121,"cited_by_count":103,"oa_url":190,"card":191,"direction":110,"ingested_from":111},"W7222603139",[165,167,169,171,173,175,177,179,182,185,188],{"name":166,"orcid":9},"Shun He",{"name":168,"orcid":9},"Sifan Li",{"name":170,"orcid":9},"Xinjian Li",{"name":172,"orcid":9},"Qianqian Luo",{"name":174,"orcid":9},"Dong Sufen",{"name":176,"orcid":9},"Lu Lei",{"name":178,"orcid":9},"Xingjiang Chen",{"name":180,"orcid":181},"Liuti Cai","https:\u002F\u002Forcid.org\u002F0000-0002-9151-8002",{"name":183,"orcid":184},"Jian‐Yu Meng","https:\u002F\u002Forcid.org\u002F0000-0002-5796-1086",{"name":186,"orcid":187},"Ning Lu","https:\u002F\u002Forcid.org\u002F0000-0003-0214-5038",{"name":189,"orcid":9},"Hancheng Wang","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1871587\u002Fpdf",{"tldr":192,"method":193,"finding":194,"direction":108,"opportunity":195},"提出MSFANet多尺度特征聚合网络，实现复杂田间背景下烟草害虫的自动识别。","多尺度金字塔池化MSPP、分层特征融合HFFB、辅助分类头与Focal Loss","模型准确率97.63%、参数量仅1.21M，在随机初始化模型中精度最高且参数最少。","可探索轻量多尺度网络在边缘设备部署及跨作物害虫迁移识别，结合类不平衡与复杂背景的鲁棒性研究。","2026-10-17T23:30:08.695455Z",{"id":198,"title":199,"url":200,"summary":201,"summary_zh":202,"content":9,"source_name":203,"source_url":200,"published_at":204,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":205,"score_detail":206,"sources":208,"tags":210,"primary_topic":212,"secondary_topics":213,"classification":219,"search_phrases":254,"slug":257,"view_count":21,"doi":258,"paper":259,"created_at":272},4660,"IoT-Based Automatic Cotton Leaf Disease Detection, Infection Percentage Estimation, and Treatment Recommendation Using Raspberry PI","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23280692","Cotton is one of the most important commercial crops worldwide, and its productivity is significantly affected by various leaf diseases such as bacterial blight, alter aria leaf spot, Cercospora leaf spot, powdery mildew, and leaf curl disease. Early and accurate identification of these diseases is essential to minimize crop losses and improve yield. Traditional disease diagnosis relies on manual field inspection by experts, which is time-consuming, labour-intensive, and often impractical for large agricultural fields. The proposed system integrates a Raspberry Pi5, a high-resolution camera module, environmental sensors, and wireless communication technologies to create an intelligent crop monitoring platform. Images of cotton leaves are captured using the Raspberry Pi camera and processed through computer vision techniques for leaf segmentation and feature extraction. A lightweight deep learning model based on MobileNetV2 is employed to automatically identify different cotton leaf diseases. In addition to disease classification, image segmentation techniques are used to calculate the percentage of infected leaf area, enabling the system to determine the severity level of infection. The system also collects environmental parameters such as temperature, humidity, soil moisture, and light intensity using IoT sensors. These parameters, along with disease diagnosis results, are transmitted to a cloud-based dashboard for real-time monitoring and analysis. Automated alerts can be sent to farmers through SMS or mobile notifications when severe infections are detected. Experimental evaluation is expected to achieve disease classification accuracy above 93%, infection area estimation error below 5%, and real-time inference within one second per image on the Raspberry Pi platform. The proposed solution offers a cost-effective, portable, and scalable approach for precision agriculture by combining artificial intelligence, image processing, and IoT technologies. The system enables early disease detection, accurate infection assessment, timely treatment recommendations, reduced pesticide usage, and improved cotton crop productivity, making it a valuable tool for smart farming applications.","棉花是全球最重要的经济作物之一，其产量受到多种叶部病害的显著影响，如细菌性枯萎病（bacterial blight）、链格孢叶斑病（alternaria leaf spot）、尾孢叶斑病（Cercospora leaf spot）、白粉病（powdery mildew）和卷叶病（leaf curl disease）。早期准确识别这些病害对于减少作物损失和提高产量至关重要。传统的病害诊断依赖专家进行人工田间巡检，这种方式耗时、费力，且对于大面积农田往往不切实际。所提出的系统集成了Raspberry Pi5、高分辨率摄像头模块、环境传感器和无线通信技术，构建了一个智能作物监测平台。利用Raspberry Pi摄像头采集棉花叶片图像，并通过计算机视觉技术进行叶片分割和特征提取。采用基于MobileNetV2的轻量级深度学习模型自动识别不同的棉花叶部病害。除病害分类外，还利用图像分割技术计算受感染叶面积百分比，使系统能够确定感染的严重程度。该系统还通过物联网传感器采集温度、湿度、土壤水分和光照强度等环境参数。这些参数连同病害诊断结果一起传输至基于云端的仪表板，用于实时监测和分析。当检测到严重感染时，可通过短信或移动通知向农民发送自动警报。实验评估预期在Raspberry Pi平台上实现高于93%的病害分类准确率、低于5%的感染面积估计误差，以及每张图像一秒以内的实时推理。所提出的解决方案结合了人工智能、图像处理和物联网技术，为精准农业提供了一种经济高效、便携且可扩展的方法。该系统能够实现病害早期检测、准确感染评估、及时治疗建议、减少农药使用并提高棉花作物产量，使其成为智慧农业应用中的一种有价值的工具。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-10T00:00:00Z",74,{"impact":125,"substance":18,"depth":19,"authority":17,"freshness":128,"relevant":21,"comment":207},"基于树莓派与轻量深度学习的棉花叶病检测与物联网监测方案，方法具体、指标明确，对智慧植保有参考价值。",[209],{"name":203,"url":200},[26,27,76,29,211],"棉花病害",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[214,218],{"code":72,"slug":215,"name":216,"path":217,"parent_code":43},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":220},{"candidates":221,"agricultural_evidence":52},[222,226,230,235,241,247],{"code":72,"score":142,"supported":52,"title":223,"abstract":224,"tags":225},[77],[76,77],[76],{"code":32,"score":59,"supported":52,"title":227,"abstract":228,"tags":229},[61],[63,61],[],{"code":66,"score":231,"supported":52,"title":232,"abstract":233,"tags":234},11,[],[26,29,69],[26,29],{"code":39,"score":13,"supported":52,"title":236,"abstract":237,"tags":240},[],[28,238,54,56,239],"计算机视觉","computer vision",[27],{"code":87,"score":73,"supported":52,"title":242,"abstract":243,"tags":246},[],[244,245],"农药","pesticide",[],{"code":248,"score":73,"supported":52,"title":249,"abstract":250,"tags":253},"H2",[],[251,252],"土壤","soil",[],[255,256],"Raspberry Pi 棉花病害检测","MobileNetV2 棉花叶病识别","RaspberryPi棉花病害检测-4660","10.5281\u002Fzenodo.23280692",{"doi":258,"openalex_id":260,"authors":261,"venue":203,"cited_by_count":103,"oa_url":200,"card":267,"direction":110,"ingested_from":111},"W7221340722",[262,264],{"name":263,"orcid":9},"Deepak R Khadse",{"name":265,"orcid":266},"Pankaj Hiraman Zope","https:\u002F\u002Forcid.org\u002F0000-0003-3669-1440",{"tldr":268,"method":269,"finding":270,"direction":110,"opportunity":271},"基于树莓派与物联网，实现棉花叶病自动检测、感染率估算与防治推荐。","树莓派5+摄像头+环境传感器，MobileNetV2分类与图像分割，云端看板。","预期分类准确率超93%，感染面积误差低于5%，单图推理1秒内。","可探索边缘端轻量模型与多病害严重度分级、精准施药决策的闭环验证。","2026-10-11T23:30:07.876353Z",{"id":274,"title":275,"url":276,"summary":277,"summary_zh":9,"content":9,"source_name":278,"source_url":276,"published_at":279,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":280,"score_detail":281,"sources":283,"tags":285,"primary_topic":286,"secondary_topics":287,"classification":289,"search_phrases":310,"slug":313,"view_count":73,"doi":314,"paper":315,"created_at":335},3463,"Cotton leaf disease classification using deep learning models for smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs40066-026-00610-2","Cotton leaf disease classification using deep learning models for smart agriculture。Agriculture & Food Security","Agriculture & Food Security","2026-09-25T00:00:00Z",66,{"impact":17,"substance":125,"depth":142,"authority":127,"freshness":13,"relevant":21,"comment":282},"核心期刊论文，方法有一定参考价值，但属细分技术研究，产业影响有限。",[284],{"name":278,"url":276},[26,27,28,211],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[288],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":290},{"candidates":291,"agricultural_evidence":52},[292,296,300,304],{"code":39,"score":127,"supported":52,"title":293,"abstract":294,"tags":295},[54],[54],[27,28],{"code":66,"score":17,"supported":52,"title":297,"abstract":298,"tags":299},[155],[155],[26],{"code":32,"score":231,"supported":52,"title":301,"abstract":302,"tags":303},[61],[61],[],{"code":305,"score":81,"supported":52,"title":306,"abstract":307,"tags":309},"A8",[],[308],"food security",[],[311,312],"棉花叶部病害 深度学习 分类","农业人工智能 智慧农业 棉花病害 深度学习","棉花叶部病害深度学习分类-3463","10.1186\u002Fs40066-026-00610-2",{"doi":314,"openalex_id":316,"authors":317,"venue":278,"cited_by_count":103,"oa_url":276,"card":9,"direction":110,"ingested_from":111},"W7214238516",[318,320,323,326,329,332],{"name":319,"orcid":9},"Hina Kiran Abbas",{"name":321,"orcid":322},"Muhammad Farrukh Shahid","https:\u002F\u002Forcid.org\u002F0009-0004-8787-1868",{"name":324,"orcid":325},"Rehab Bahaaddin Ashari","https:\u002F\u002Forcid.org\u002F0000-0003-1225-7535",{"name":327,"orcid":328},"Arwa Mashat","https:\u002F\u002Forcid.org\u002F0000-0002-0612-6005",{"name":330,"orcid":331},"Tariq Jamil Saifullah Khanzada","https:\u002F\u002Forcid.org\u002F0000-0003-1617-4403",{"name":333,"orcid":334},"M. Hassan Tanveer","https:\u002F\u002Forcid.org\u002F0000-0001-9266-6368","2026-09-25T23:30:09.012501Z",{"id":337,"title":338,"url":339,"summary":340,"summary_zh":9,"content":9,"source_name":341,"source_url":9,"published_at":122,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":342,"sources":344,"tags":346,"primary_topic":349,"secondary_topics":350,"classification":352,"search_phrases":374,"slug":377,"view_count":103,"doi":9,"paper":378,"created_at":387},5532,"精准棉花病害管理：融合无人机影像与定制YOLO深度学习框架","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24182","研究提出单阶段全卷积检测器Cotton-Vision，融合幽灵瓶颈通道注意力与跨阶段部分融合三项轻量化设计，参数减少20%。在巴基斯坦旁遮普省7个棉田采集的2700余幅高分辨率无人机影像上测试，精确率97.7%、召回率77.7%、mAP@50为85.6%，较主流基线提升10至20个百分点。预印本，仅在巴基斯坦单一区域验证。","arXiv 预印本 2026-10-16",{"impact":125,"substance":126,"depth":19,"authority":128,"freshness":73,"relevant":21,"comment":343},"方法轻量化与实测数据有增量，但仅单区域预印本验证，属细分技术进展，可入精选备选。",[345],{"name":341,"url":339},[26,27,347,348,211],"目标检测","无人机遥感",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[351],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":353},{"candidates":354,"agricultural_evidence":52},[355,361,365,369],{"code":39,"score":356,"supported":52,"title":357,"abstract":359,"tags":360},18,[28,358],"YOLO",[],[27,347],{"code":32,"score":231,"supported":52,"title":362,"abstract":363,"tags":364},[63],[61],[],{"code":66,"score":21,"supported":14,"title":366,"abstract":367,"tags":368},[],[],[26],{"code":370,"score":21,"supported":14,"title":371,"abstract":372,"tags":373},"I3",[],[],[348],[375,376],"Cotton-Vision 棉花病害","无人机影像 棉花 检测","Cotton-Vision棉花病害-5532",{"doi":9,"openalex_id":9,"authors":379,"venue":9,"cited_by_count":103,"oa_url":9,"card":380,"direction":384,"ingested_from":386},[],{"tldr":381,"method":382,"finding":383,"direction":384,"opportunity":385},"提出轻量化检测器Cotton-Vision，用无人机影像精准识别棉花病害。","基于YOLO改进，融合幽灵瓶颈通道注意力与跨阶段部分融合，用2700余幅无人机影","精确率97.7%、mAP@50达85.6%，较主流基线提升10至20个百分点。","农业遥感与作物表型","仅在巴基斯坦单一区域验证，可扩展至多区域多品种，并探索边缘设备实时部署。","agent","2026-10-22T00:07:09.816790Z",{"id":389,"title":390,"url":391,"summary":392,"summary_zh":9,"content":9,"source_name":393,"source_url":9,"published_at":9,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":394,"score_detail":395,"sources":398,"tags":400,"primary_topic":403,"secondary_topics":404,"classification":406,"search_phrases":428,"slug":431,"view_count":103,"doi":9,"paper":432,"created_at":439},4966,"基于改进YOLO 11n模型的棉花田间复杂环境障碍物检测方法","http:\u002F\u002Fwww.qikanzj.com\u002Fhek\u002Fnyjxxb\u002Fmulu\u002F455619.html","韩科立、王振坤等(中国农业机械化科学研究院集团有限公司;现代农装科技股份有限公司;农业装备技术全国重点实验室)针对棉花田间复杂环境障碍物被遮挡致准确检测难、边缘设备算力有限的问题,提出一种基于改进YOLO 11n模型的田间障碍物检测方法。采用StarNet作为主要特征提取网络并引入DBA模块增强多尺度特征交互能力;使用KAGNConv替换C3k2瓶颈结构增强灵活性与可解释性;集成分离与增强注意力模块SEAM至检测头增强遮挡场景能力。试验结果表明,改进模型YOLO 11n-SKS精确率、召回率、mAP50、mAP50-95分别提升2.3、2.1、1.3、1.4个百分点,达到91.7%、88.3%、91.9%、62.3%。","《农业机械学报》",71,{"impact":125,"substance":126,"depth":356,"authority":59,"freshness":396,"relevant":21,"comment":397},2,"改进YOLO 11n实现棉花田间遮挡障碍物检测，方法新颖、指标提升明确，属农业人工智能细分领域实质进展，但发布日期未知、时效性偏弱。",[399],{"name":393,"url":391},[26,27,63,401,402],"障碍物检测","采棉机",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[405],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":407},{"candidates":408,"agricultural_evidence":52},[409,413,417,424],{"code":32,"score":17,"supported":52,"title":410,"abstract":411,"tags":412},[63],[63],[63],{"code":39,"score":17,"supported":52,"title":414,"abstract":415,"tags":416},[358],[358],[27],{"code":418,"score":73,"supported":52,"title":419,"abstract":420,"tags":423},"F5",[],[421,422],"农业机械","机械化",[],{"code":66,"score":21,"supported":14,"title":425,"abstract":426,"tags":427},[],[],[26],[429,430],"中国农业机械化科学研究院 采棉机 障碍物检测","YOLO 11n 棉花 田间障碍物","中国农业机械化科学研究院采棉机障碍物检测-4966",{"doi":9,"openalex_id":9,"authors":433,"venue":9,"cited_by_count":103,"oa_url":9,"card":434,"direction":108,"ingested_from":386},[],{"tldr":435,"method":436,"finding":437,"direction":108,"opportunity":438},"提出改进YOLO 11n模型，实现棉花田间复杂遮挡环境下障碍物高精度检测。","以StarNet为骨干，引入DBA、KAGNConv与SEAM注意力模块，改进Y","改进模型精确率91.7%、召回率88.3%、mAP50达91.9%，各项指标均提升。","可探索轻量化模型在边缘设备实时部署，并扩展至多作物多类障碍物检测。","2026-10-15T00:06:08.577388Z",{"id":441,"title":442,"url":443,"summary":444,"summary_zh":445,"content":9,"source_name":446,"source_url":443,"published_at":447,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":448,"score_detail":449,"sources":451,"tags":453,"primary_topic":456,"secondary_topics":457,"classification":459,"search_phrases":484,"slug":487,"view_count":81,"doi":488,"paper":489,"created_at":519},2288,"Field-based deep learning classification of cotton and weeds for machine-vision-assisted intra-row weed management","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102567","Timely and precise weed management is essential for reducing crop-weed competition, labour requirements, and unnecessary weed-control inputs in cotton production. Reliable discrimination between cotton plants and weeds is a prerequisite for automated intra-row weed management. This study evaluated two pretrained convolutional neural network models, MobileNetV2 and DenseNet169, for binary classification of cotton and weeds using field-acquired RGB images. Images were collected from a farmer-managed cotton field in Haryana and an experimental cotton field at ICAR-Indian Agricultural Research Institute, New Delhi, during May-September 2025 under natural field conditions. The dataset comprised 2,195 weed images representing six predominant weed species and 1,211 cotton images. Images were classified into two operational classes, cotton and weed. Training images were augmented to address the original class imbalance, whereas validation and test images were retained without augmentation. Transfer learning was used to fine-tune both models, and performance was evaluated using accuracy, precision, recall, F1-score, confusion matrices, and inference time. Both models achieved high classification performance. DenseNet169 attained a test accuracy of 99.47%, compared with 98.27% for MobileNetV2. The results indicate that pretrained CNNs can provide accurate image-level discrimination between cotton and weed vegetation under the investigated field conditions. DenseNet169 therefore shows promise as a visual-perception component for machine-vision-assisted intra-row weed management. However, the present study is limited to image-level classification and does not demonstrate plant localization, continuous machine operation, actuator coordination, or field-scale weed removal. Further validation under independent locations, varying weed densities, illumination conditions, machine travel speeds, and complete perception-decision-actuation pipelines is required.","及时、精准的杂草管理对于减少棉花生产中的棉草竞争、劳动力需求和不必要的除草投入至关重要。可靠地区分棉花植株与杂草是实现行内自动化杂草管理的前提。本研究评估了两种预训练卷积神经网络模型——MobileNetV2和DenseNet169，利用田间采集的RGB图像对棉花与杂草进行二分类。图像于2025年5月至9月期间，在自然田间条件下，采集自哈里亚纳邦一处农民管理的棉田以及新德里ICAR-印度农业研究所的实验棉田。数据集包含2，195幅杂草图像，涵盖六种主要杂草种类，以及1，211幅棉花图像。图像被分为棉花和杂草两个操作类别。训练图像经过增强处理以解决原始类别不平衡问题，而验证和测试图像则保留未增强状态。采用迁移学习对两种模型进行微调，并使用准确率、精确率、召回率、F1分数、混淆矩阵和推理时间评估性能。两种模型均取得了较高的分类性能。DenseNet169的测试准确率达到99.47%，而MobileNetV2为98.27%。结果表明，预训练卷积神经网络能够在所研究的田间条件下实现棉花与杂草植被在图像层面的准确区分。因此，DenseNet169有望作为机器视觉辅助行内杂草管理的视觉感知组件。然而，本研究仅限于图像层面的分类，并未展示植株定位、机器连续作业、执行器协调或田间规模除草。仍需在独立地点、不同杂草密度、光照条件、机器行进速度以及完整的感知-决策-执行流程下进行进一步验证。","Smart Agricultural Technology","2026-09-11T00:00:00Z",67,{"impact":17,"substance":356,"depth":125,"authority":127,"freshness":20,"relevant":21,"comment":450},"基于田间RGB图像的棉花与杂草深度学习分类研究，DenseNet169测试准确率达99.47%，方法扎实但仅限图像级分类，尚缺定位与执行环节验证，属细分领域技术进展。",[452],{"name":446,"url":443},[26,27,63,454,455],"杂草识别","机器视觉",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[458],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},{"status":45,"locked":14,"version":46,"method":47,"evidence":460},{"candidates":461,"agricultural_evidence":52},[462,466,470,476,480],{"code":32,"score":142,"supported":52,"title":463,"abstract":464,"tags":465},[61],[63,61],[63],{"code":39,"score":13,"supported":52,"title":467,"abstract":468,"tags":469},[54],[],[27,455],{"code":87,"score":471,"supported":52,"title":472,"abstract":473,"tags":475},4,[],[474],"除草",[454],{"code":80,"score":81,"supported":52,"title":477,"abstract":478,"tags":479},[],[84],[],{"code":66,"score":21,"supported":14,"title":481,"abstract":482,"tags":483},[],[],[26],[485,486],"农业人工智能 智慧农业 机器视觉 杂草识别","农业人工智能 智慧农业","农业人工智能智慧农业机器视觉杂草识别-2288","10.1016\u002Fj.atech.2026.102567",{"doi":488,"openalex_id":490,"authors":491,"venue":446,"cited_by_count":103,"oa_url":443,"card":514,"direction":108,"ingested_from":111},"W7212315230",[492,494,497,500,503,505,508,510,512],{"name":493,"orcid":9},"Shaik Nasreen",{"name":495,"orcid":496},"Roaf Ahmad Parray","https:\u002F\u002Forcid.org\u002F0000-0002-8303-1990",{"name":498,"orcid":499},"Parveen Dhanger","https:\u002F\u002Forcid.org\u002F0000-0001-5250-0275",{"name":501,"orcid":502},"Rishi Raj","https:\u002F\u002Forcid.org\u002F0009-0002-7918-1141",{"name":504,"orcid":9},"Tapan Kumar Khura",{"name":506,"orcid":507},"P. Sahoo","https:\u002F\u002Forcid.org\u002F0000-0002-3888-8506",{"name":509,"orcid":9},"Tushar Dhar",{"name":511,"orcid":9},"Prajwal R",{"name":513,"orcid":9},"Sripriyanka S. Nalla",{"tldr":515,"method":516,"finding":517,"direction":108,"opportunity":518},"用MobileNetV2和DenseNet169对田间棉花与杂草图像做二分类，验证机器视觉除草可行性","迁移学习微调两种预训练CNN，使用2025年田间RGB图像共3406张。","DenseNet169测试准确率达99.47%，优于MobileNetV2的98.27%。","可延伸至植株定位、多光照与密度条件下的感知-决策-执行全流程田间验证。","2026-09-13T23:30:04.183782Z"]