[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-5302":3,"related-5302":66},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":23,"tags":25,"primary_topic":8,"secondary_topics":30,"classification":31,"search_phrases":50,"slug":53,"view_count":54,"doi":8,"paper":55,"created_at":65},5302,"Peach-YOLO:基于改进YOLOv8与多注意力融合的复杂环境桃实时检测方法","https:\u002F\u002Fm2.mtmt.hu\u002Fapi\u002Fpublication\u002F37332430","Frontiers in Plant Science 2026年第17期刊载You, Shuanghong等的Peach-YOLO论文,基于改进YOLOv8n框架,引入RFAConv模块强化特征表示、融合CAFM模块提取关键果实特征、采用CA-HSFPN金字塔网络、引入Shape-IoU损失函数;在自建桃数据集上模型仅5.0 MB,推理速度115.7 FPS,mAP@0.5达82.2%、精度78.9%、召回率76.4%,较YOLOv8n基线mAP、精度、召回率分别提升3.0、3.6、4.8个百分点。",null,"Frontiers in Plant Science(2026, 17)","2026-10-16T00:00:00Z","论文",10,false,72,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},12,21,17,13,9,1,"方法改进扎实、指标明确，但属细分作物检测算法研究，产业影响有限，适合作为技术类精选。",[24],{"name":9,"url":6},[26,27,28,29],"智慧农业","农业人工智能","目标检测","果实识别",[],{"status":32,"locked":13,"version":33,"method":34,"evidence":35},"needs_review","2026-10-22.2","rules",{"candidates":36,"agricultural_evidence":13,"reason":49},[37,44],{"code":38,"score":19,"supported":39,"title":40,"abstract":42,"tags":43},"I2",true,[41],"YOLO",[41],[27,28],{"code":45,"score":21,"supported":13,"title":46,"abstract":47,"tags":48},"I1",[],[],[26],"农业相关性需要复核",[51,52],"Peach-YOLO 桃检测","YOLOv8 果实检测","Peach-YOLO桃检测-5302",2,{"doi":8,"openalex_id":8,"authors":56,"venue":8,"cited_by_count":57,"oa_url":8,"card":58,"direction":62,"ingested_from":64},[],0,{"tldr":59,"method":60,"finding":61,"direction":62,"opportunity":63},"提出Peach-YOLO改进YOLOv8n模型，实现复杂环境下桃果实实时检测。","改进YOLOv8n，引入RFAConv、CAFM、CA-HSFPN与Shape-","模型仅5.0MB、115.7FPS，mAP@0.5达82.2%，较基线提升3.0个百分点。","农业遥感与作物表型","可探索轻量检测模型在采摘机器人边缘端部署及多果园场景泛化能力。","agent","2026-10-19T00:07:48.648980Z",{"total":67,"page":21,"page_size":67,"items":68},6,[69,134,204,269,339,384],{"id":70,"title":71,"url":72,"summary":73,"summary_zh":8,"content":8,"source_name":74,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":75,"score_detail":76,"sources":79,"tags":81,"primary_topic":84,"secondary_topics":90,"classification":96,"search_phrases":122,"slug":125,"view_count":57,"doi":8,"paper":126,"created_at":133},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",69,{"impact":77,"substance":17,"depth":18,"authority":20,"freshness":67,"relevant":21,"comment":78},16,"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。",[80],{"name":74,"url":72},[26,27,28,82,83],"无人机遥感","棉花病害",{"code":85,"slug":86,"name":87,"path":88,"parent_code":89},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops","A",[91],{"code":38,"slug":92,"name":93,"path":94,"parent_code":95},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":97,"locked":13,"version":33,"method":34,"evidence":98},"classified",{"candidates":99,"agricultural_evidence":39},[100,106,113,117],{"code":38,"score":101,"supported":39,"title":102,"abstract":104,"tags":105},18,[103,41],"深度学习",[],[27,28],{"code":85,"score":107,"supported":39,"title":108,"abstract":110,"tags":112},11,[109],"棉花",[111],"cotton",[],{"code":45,"score":21,"supported":13,"title":114,"abstract":115,"tags":116},[],[],[26],{"code":118,"score":21,"supported":13,"title":119,"abstract":120,"tags":121},"I3",[],[],[82],[123,124],"Cotton-Vision 棉花病害","无人机影像 棉花 检测","Cotton-Vision棉花病害-5532",{"doi":8,"openalex_id":8,"authors":127,"venue":8,"cited_by_count":57,"oa_url":8,"card":128,"direction":62,"ingested_from":64},[],{"tldr":129,"method":130,"finding":131,"direction":62,"opportunity":132},"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。","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。","2026-10-22T00:07:09.816790Z",{"id":135,"title":136,"url":137,"summary":138,"summary_zh":139,"content":8,"source_name":140,"source_url":137,"published_at":141,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":142,"score_detail":143,"sources":147,"tags":149,"primary_topic":151,"secondary_topics":152,"classification":153,"search_phrases":173,"slug":176,"view_count":57,"doi":177,"paper":178,"created_at":203},5393,"IMPLEMENTATION OF META-HEURISTIC OPTIMIZATION ALGORITHMS IN CONVOLUTIONAL NEURAL NETWORKS FOR OBJECT DETECTION AND CLASSIFICATION","https:\u002F\u002Fdoi.org\u002F10.22190\u002Ffuacr251214006i","CNNs have grown to form the backbone of modern computer vision systems, producing outstanding object detection and classification performance within diverse application domains. However, the manual development of suitable CNN architectures and, correspondingly, the selection of optimal hyperparameter sets are still inconvenient and time-consuming processes. The aim of this survey is to contribute a comprehensive review regarding how meta-heuristic optimization algorithms such as Genetic Algorithms, Particle Swarm Optimization, and other swarm intelligence-based methods are embedded in the CNN-based pipelines with the intent to achieve better detection and classification results. We gather recent progress regarding automated architecture search, hyperparameter tuning, model compression, and ensemble learning. We identify how hybrid optimization approaches consistently yield superior accuracy with computational efficiency by systematically reviewing applications in medical imaging, agriculture, vehicular recognition, and remote sensing. Optimized CNN models outperform their baseline versions many times, especially in health diagnostics, where several fine-tuned architectures reach accuracies greater than 99%. This survey provides a structured and updated overview of methodological developments, implementation practices, and emerging research directions for those practitioners and researchers who will apply meta-heuristic optimization within deep learning-based computer vision systems.","卷积神经网络（CNNs）已发展成为现代计算机视觉系统的核心骨干，在多个应用领域中展现出卓越的目标检测与分类性能。然而，人工设计合适的CNN架构并相应地选择最优超参数集仍然是繁琐且耗时的过程。本综述旨在全面回顾遗传算法、粒子群优化及其他群体智能方法等元启发式优化算法如何嵌入基于CNN的流程中以获得更优的检测与分类结果。我们汇集了自动化架构搜索、超参数调优、模型压缩和集成学习方面的最新进展。通过系统梳理在医学影像、农业、车辆识别和遥感领域的应用，我们揭示了混合优化方法如何以计算效率持续实现更优的精度。优化后的CNN模型在多数情况下远超其基线版本，尤其在健康诊断领域，若干精细调优的架构达到了超过99%的准确率。本综述为将在基于深度学习的计算机视觉系统中应用元启发式优化的从业者和研究者，提供了关于方法学进展、实施实践和新兴研究方向的结构化且更新的概述。","Facta Universitatis Series Automatic Control and Robotics","2026-10-18T00:00:00Z",61,{"impact":144,"substance":101,"depth":145,"authority":16,"freshness":144,"relevant":21,"comment":146},8,15,"综述系统梳理元启发式优化在CNN视觉任务中的嵌入方法，含农业应用，但属通用方法综述，非农业专属突破。",[148],{"name":140,"url":137},[26,27,28,150],"遥感",{"code":38,"slug":92,"name":93,"path":94,"parent_code":95},[],{"status":97,"locked":13,"version":33,"method":34,"evidence":154},{"candidates":155,"agricultural_evidence":39},[156,163,169],{"code":38,"score":107,"supported":39,"title":157,"abstract":158,"tags":162},[],[28,103,159,160,161],"计算机视觉","deep learning","computer vision",[27,28],{"code":118,"score":164,"supported":39,"title":165,"abstract":166,"tags":168},7,[],[150,167],"remote sensing",[150],{"code":45,"score":21,"supported":13,"title":170,"abstract":171,"tags":172},[],[],[26],[174,175],"元启发式优化 卷积神经网络","CNN 目标检测","元启发式优化卷积神经网络-5393","10.22190\u002Ffuacr251214006i",{"doi":177,"openalex_id":179,"authors":180,"venue":140,"cited_by_count":57,"oa_url":137,"card":196,"direction":62,"ingested_from":202},"W7222920697",[181,184,187,190,193],{"name":182,"orcid":183},"Nikola Ivačko","https:\u002F\u002Forcid.org\u002F0000-0003-4367-119X",{"name":185,"orcid":186},"Ivan S Ciric","https:\u002F\u002Forcid.org\u002F0000-0003-0430-8937",{"name":188,"orcid":189},"Milica Tasić","https:\u002F\u002Forcid.org\u002F0009-0001-9160-6379",{"name":191,"orcid":192},"Ljiljana Radović","https:\u002F\u002Forcid.org\u002F0000-0002-4099-1289",{"name":194,"orcid":195},"Žarko Ćojbašić","https:\u002F\u002Forcid.org\u002F0000-0002-4581-1048",{"tldr":197,"method":198,"finding":199,"direction":200,"opportunity":201},"综述元启发式算法嵌入CNN流程以优化目标检测与分类的研究进展。","系统综述遗传算法、粒子群等群智能方法用于CNN架构搜索与超参调优。","混合优化方法在医学影像、农业等领域兼顾精度与效率，部分模型准确率超99%。","农业人工智能与决策模型","农业场景下元启发式优化CNN的架构搜索与轻量化部署仍缺乏系统基准与田间验证。","openalex","2026-10-20T23:30:19.233167Z",{"id":205,"title":206,"url":207,"summary":208,"summary_zh":209,"content":8,"source_name":210,"source_url":207,"published_at":211,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":212,"score_detail":213,"sources":215,"tags":217,"primary_topic":8,"secondary_topics":220,"classification":221,"search_phrases":232,"slug":235,"view_count":21,"doi":236,"paper":237,"created_at":268},5136,"Enhanced YOLOv8n for adhered leaf cluster detection in RGB-based fresh green tea leaf sorting","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102640","High-precision detection of adhered leaf clusters (ALCs) based on RGB imaging has long posed a critical challenge in the post-harvest sorting of fresh green tea leaves. To address this issue, this study designed an enhanced YOLOv8n model, named SDSI-YOLO, by integrating the StarNet backbone, DySample upsampling method, and Shape-Inner-IoU loss function to improve ALC detection performance. Experimental results demonstrated that SDSI-YOLO achieved an optimal balance between accuracy and efficiency, with a mean average precision of 92.6% and an F 1 -score of 89.2. It also maintained lightweight characteristics, with a model size of only 3.89 MB and efficient inference speed of 3.1 ms per frame. Ablation experiments confirmed the effectiveness of each module in constructing SDSI-YOLO. The results of visualization tests indicated that the model exhibited a stronger capability in capturing critical details and structural features of ALCs. This research provides a feasible solution for the industrial sorting of fresh green tea leaves, demonstrating considerable application potential.","基于RGB成像的高精度粘连叶簇（adhered leaf clusters，ALCs）检测一直是新鲜绿茶鲜叶采后分选中的关键难题。为解决该问题，本研究通过融合StarNet主干网络、DySample上采样方法和Shape-Inner-IoU损失函数，设计了一种增强型YOLOv8n模型，命名为SDSI-YOLO，以提升ALCs检测性能。实验结果表明，SDSI-YOLO在精度与效率之间实现了最优平衡，平均精度均值达到92.6%，F1分数为89.2。同时，该模型保持了轻量化特性，模型大小仅为3.89 MB，推理速度高达每帧3.1 ms。消融实验验证了各模块在构建SDSI-YOLO中的有效性。可视化测试结果表明，该模型在捕捉ALCs关键细节和结构特征方面表现出更强的能力。本研究为新鲜绿茶鲜叶的工业化分选提供了一种可行方案，具有相当大的应用潜力。","Smart Agricultural Technology","2026-10-14T00:00:00Z",76,{"impact":77,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":214},"提出轻量化改进YOLOv8n模型用于鲜茶叶粘连叶簇检测，指标与推理速度均有实测数据，方法新颖且面向产业分选场景，具备精选价值。",[216],{"name":210,"url":207},[26,27,28,218,219],"YOLOv8","茶叶分选",[],{"status":32,"locked":13,"version":33,"method":34,"evidence":222},{"candidates":223,"agricultural_evidence":13,"reason":49},[224,228],{"code":38,"score":67,"supported":39,"title":225,"abstract":226,"tags":227},[],[41],[218,27,28],{"code":45,"score":21,"supported":13,"title":229,"abstract":230,"tags":231},[],[],[26],[233,234],"SDSI-YOLO 茶叶分选","鲜叶 粘连叶簇 检测","SDSI-YOLO茶叶分选-5136","10.1016\u002Fj.atech.2026.102640",{"doi":236,"openalex_id":238,"authors":239,"venue":210,"cited_by_count":57,"oa_url":207,"card":263,"direction":200,"ingested_from":202},"W7221817030",[240,242,244,246,248,251,253,256,258,260],{"name":241,"orcid":8},"Jian Wang",{"name":243,"orcid":8},"Yi-jian Liu",{"name":245,"orcid":8},"Yin-yin Lin",{"name":247,"orcid":8},"Cheng-chen Zhang",{"name":249,"orcid":250},"Shengnan Liu","https:\u002F\u002Forcid.org\u002F0000-0001-9886-068X",{"name":252,"orcid":8},"Bin Ji",{"name":254,"orcid":255},"Qiu Wang","https:\u002F\u002Forcid.org\u002F0000-0002-5861-4115",{"name":257,"orcid":8},"Li-hong Du",{"name":259,"orcid":8},"Lei Zheng",{"name":261,"orcid":262},"Fei Ma","https:\u002F\u002Forcid.org\u002F0000-0002-7313-1787",{"tldr":264,"method":265,"finding":266,"direction":200,"opportunity":267},"提出改进YOLOv8n模型SDSI-YOLO，实现鲜茶叶粘连叶簇的高精度检测。","融合StarNet主干、DySample上采样与Shape-Inner-IoU损","mAP达92.6%、F1为89.2，模型仅3.89MB、单帧3.1ms，精度与效率兼顾。","可探索轻量检测模型在茶叶分选产线的嵌入式部署与多品类叶簇泛化能力。","2026-10-17T23:30:03.404999Z",{"id":270,"title":271,"url":272,"summary":273,"summary_zh":274,"content":8,"source_name":210,"source_url":272,"published_at":275,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":276,"score_detail":277,"sources":281,"tags":283,"primary_topic":286,"secondary_topics":287,"classification":294,"search_phrases":312,"slug":315,"view_count":21,"doi":316,"paper":317,"created_at":338},4975,"Detection of mounting behavior and mounted ewes in group-housed environments using HSM-YOLO algorithm","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102639","The detection of mounting behavior and the mounted ewe is critical for estrus monitoring in sheep. However, existing methods suffer from low detection efficiency and high false detection rates due to the similarity between non-sexual and sexual mounting behaviors and the complexity of group-housed environments. To address these challenges, this study introduced a fine-grained classification of mounting behaviors into sexual and non-sexual categories, and further proposed an improved detection algorithm named Hu Sheep Mounting-YOLO (HSM-YOLO) based on the YOLOv11n architecture for detecting sheep mounting behavior and the corresponding mounted ewes. The HSM-YOLO algorithm employed depthwise separable convolution (DWSConv) to reduce the model complexity. To improve behavioral feature extraction and pose discrimination capabilities, it further integrated the C3k2_RepViTBlock-RepBottleneck (C3k2_BiRep) module, an efficient multi-scale attention (EMA) module, and the Inner-SIoU loss function. To evaluate the performance of the proposed algorithm, comparative experiments were conducted on a multiple scenarios dataset. The experimental results showed that the HSM-YOLO model achieved an F1 score of 88.1%, a mean Average Precision (mAP) of 90.2%, with 2.2M parameters and 5.4G floating-point operations (FLOPs). Compared to the baseline YOLOv11n model, HSM-YOLO improved the F1 score by 5.8% and the mAP by 5.7%, while reducing both model parameters and computational complexity. When deployed and tested on an RTX 4090, the system delivered an average detection speed of 47.9 frames per second (FPS). Moreover, despite considerable variation in detection performance across the four different imaging scenarios, HSM-YOLO consistently demonstrated superior detection accuracy. These results confirm that the HSM-YOLO provides a robust and efficient solution for detecting sheep mounting behaviors and mounted ewes in complex and variable scenarios, paving the way for the development of automated estrus monitoring systems.","检测羊只的爬跨行为及被爬跨母羊对于绵羊发情监测至关重要。然而，由于非性爬跨行为与性爬跨行为之间的相似性以及群养环境的复杂性，现有方法存在检测效率低和误检率高的问题。为解决这些挑战，本研究将爬跨行为细粒度分类为性爬跨和非性爬跨两类，并进一步提出了一种基于YOLOv11n架构的改进检测算法——湖羊爬跨-YOLO（HSM-YOLO），用于检测绵羊爬跨行为及相应的被爬跨母羊。HSM-YOLO算法采用深度可分离卷积（DWSConv）以降低模型复杂度。为提高行为特征提取和姿态判别能力，该算法进一步集成了C3k2_RepViTBlock-RepBottleneck（C3k2_BiRep）模块、高效多尺度注意力（EMA）模块以及Inner-SIoU损失函数。为评估所提算法的性能，在多场景数据集上进行了对比实验。实验结果表明，HSM-YOLO模型的F1分数达到88.1%，平均精度均值（mAP）达到90.2%，参数量为2.2M，浮点运算量为5.4G。与基线YOLOv11n模型相比，HSM-YOLO的F1分数提高了5.8%，mAP提高了5.7%，同时降低了模型参数量和计算复杂度。在RTX 4090上部署测试时，系统平均检测速度达到47.9帧每秒（FPS）。此外，尽管在四种不同成像场景下检测性能存在较大差异，HSM-YOLO始终展现出优越的检测精度。这些结果证实，HSM-YOLO为复杂多变场景下检测绵羊爬跨行为和被爬跨母羊提供了一种鲁棒且高效的解决方案，为自动化发情监测系统的开发奠定了基础。","2026-10-12T00:00:00Z",81,{"impact":101,"substance":278,"depth":101,"authority":279,"freshness":20,"relevant":21,"comment":280},22,14,"提出HSM-YOLO算法实现羊只爬跨行为与受体母羊检测，为自动化发情监测提供高效方案，方法新颖、指标扎实，具行业参考价值。",[282],{"name":210,"url":272},[26,27,28,284,285],"畜牧养殖","羊只发情监测",{"code":38,"slug":92,"name":93,"path":94,"parent_code":95},[288],{"code":289,"slug":290,"name":291,"path":292,"parent_code":293},"C2","ruminants","草食牲畜","\u002Ftopic\u002Flivestock\u002Fruminants","C",{"status":97,"locked":13,"version":33,"method":34,"evidence":295},{"candidates":296,"agricultural_evidence":39},[297,301,308],{"code":38,"score":19,"supported":39,"title":298,"abstract":299,"tags":300},[41],[41],[27,28],{"code":289,"score":20,"supported":39,"title":302,"abstract":303,"tags":307},[],[304,305,306],"湖羊","绵羊","sheep",[],{"code":45,"score":21,"supported":13,"title":309,"abstract":310,"tags":311},[],[],[26],[313,314],"HSM-YOLO 羊只爬跨行为 检测","绵羊 发情监测 目标检测","HSM-YOLO羊只爬跨行为检测-4975","10.1016\u002Fj.atech.2026.102639",{"doi":316,"openalex_id":318,"authors":319,"venue":210,"cited_by_count":57,"oa_url":272,"card":333,"direction":200,"ingested_from":202},"W7221573347",[320,322,324,326,329,331],{"name":321,"orcid":8},"Yanchao Wang",{"name":323,"orcid":8},"Ming Gao",{"name":325,"orcid":8},"Zhixin Hua",{"name":327,"orcid":328},"Asim Shakeel","https:\u002F\u002Forcid.org\u002F0000-0003-2942-9081",{"name":330,"orcid":8},"Chengyu Ren",{"name":332,"orcid":8},"Kaiying Wang",{"tldr":334,"method":335,"finding":336,"direction":200,"opportunity":337},"提出HSM-YOLO算法，在群养环境下检测羊爬跨行为及被爬跨母羊，用于发情监测。","基于YOLOv11n改进，引入DWSConv、C3k2_BiRep、EMA和In","F1达88.1%、mAP达90.2%，参数量2.2M，较基线提升且更轻量，检测速度47.9 FPS。","可延伸至多物种发情行为细粒度识别与轻量化边缘部署，结合时序信息提升监测实时性。","2026-10-15T23:30:03.393878Z",{"id":340,"title":341,"url":342,"summary":343,"summary_zh":8,"content":8,"source_name":344,"source_url":8,"published_at":345,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":346,"score_detail":347,"sources":349,"tags":351,"primary_topic":353,"secondary_topics":354,"classification":355,"search_phrases":371,"slug":374,"view_count":54,"doi":375,"paper":376,"created_at":383},4643,"多尺度异构特征协同的棉花叶片病害检测模型(Multi-Scale Heterogeneous Synergistic Model for Cotton Leaf Disease Detection)","https:\u002F\u002Fwww.smartag.net.cn\u002FCN\u002F10.12133\u002Fj.smartag.SA202601027","辽宁工程技术大学沈学利团队基于改进RT-DETR提出MHSF-DETR,通过分层上下文选择主干、可学习加权融合与边界感知重构,实现多尺度异构特征协同。相比基准模型参数量与计算量分别降低22.42%和13.29%,mAP50提升3.2个百分点,外部数据集泛化mAP分别达80.3%与86.9%。局限:模型针对特定病斑尺度设计,跨作物迁移需重新训练。","智慧农业(中英文) 2026年第4期 2026-10-08","2026-10-08T00:00:00Z",78,{"impact":77,"substance":278,"depth":101,"authority":279,"freshness":144,"relevant":21,"comment":348},"改进RT-DETR的棉花叶片病害检测模型，参数量与计算量双降且精度提升，方法新颖、数据扎实，属细分领域实质进展。",[350],{"name":344,"url":342},[26,27,159,28,352],"棉花病害检测",{"code":85,"slug":86,"name":87,"path":88,"parent_code":89},[],{"status":97,"locked":13,"version":33,"method":34,"evidence":356},{"candidates":357,"agricultural_evidence":39},[358,362,367],{"code":85,"score":77,"supported":39,"title":359,"abstract":360,"tags":361},[109,111],[],[],{"code":38,"score":363,"supported":13,"title":364,"abstract":365,"tags":366},3,[],[],[27,28,159],{"code":45,"score":21,"supported":13,"title":368,"abstract":369,"tags":370},[],[],[26],[372,373],"辽宁工程技术大学 棉花叶片病害检测","MHSF-DETR 棉花病害","辽宁工程技术大学棉花叶片病害检测-4643","10.12133\u002Fj.smartag.sa202601027",{"doi":375,"openalex_id":8,"authors":377,"venue":8,"cited_by_count":57,"oa_url":8,"card":378,"direction":200,"ingested_from":64},[],{"tldr":379,"method":380,"finding":381,"direction":200,"opportunity":382},"提出MHSF-DETR棉花叶片病害检测模型，提升精度并降低参数量与计算量。","改进RT-DETR，采用分层上下文选择主干、可学习加权融合与边界感知重构。","mAP50提升3.2个百分点，参数量与计算量分别降低22.42%和13.29%，外部数据集泛化良好。","可探索跨作物病害检测的轻量化迁移学习与域自适应方法，减少重新训练成本。","2026-10-11T00:09:47.309161Z",{"id":385,"title":386,"url":387,"summary":388,"summary_zh":389,"content":8,"source_name":390,"source_url":387,"published_at":391,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":75,"score_detail":392,"sources":394,"tags":396,"primary_topic":399,"secondary_topics":403,"classification":404,"search_phrases":432,"slug":435,"view_count":57,"doi":436,"paper":437,"created_at":453},4564,"Enhancing Long-Tailed Macaque Pest Identification Using a Two-Stage YOLO-EfficientNet Pipeline","https:\u002F\u002Fdoi.org\u002F10.36456\u002F1hrtgc63","Long-tailed macaque (Macaca fascicularis) infestations have reduced corn production from a potential 7–9 tons to only 4–5 tons per growing season (±40%), while conventional deterrence methods such as loud noise and shooting have proven ineffective. This study developed an Internet of Things (IoT)-based pest repellent system using audiosonic waves, an HC-SR501 sensor, and an ESP32-CAM module, incorporating YOLOv8 and EfficientNet-B0 to replace the previously used YOLOv3 Tiny model, with the aim of improving detection accuracy and distinguishing long-tailed macaques from humans. Since population density at the partner's farmland was insufficient to construct an adequate dataset, image data were collected at Kembang Island, South Kalimantan, which serves as a natural habitat supporting a large and observable population of long-tailed macaques, thereby enabling the acquisition of visually representative and diverse training data. A total of 3,083 images were annotated using Label Studio under two classes (human and long-tailed macaque), then divided at an 80:10:10 ratio without augmentation. The YOLOv8n model was trained for 50 epochs, after which validation detections were cropped into 920 images to fine-tune EfficientNet-B0 over 10 epochs. Evaluation results show that YOLOv8 achieved a precision of 0.959, recall of 0.965, mAP50 of 0.979, and mAP50-95 of 0.723, while EfficientNet-B0 reached validation accuracies of 98.4–100% with no significant overfitting. The two-stage detection-classification pipeline was successfully tested end-to-end on 121 field images, demonstrating substantial performance improvements over the prior model and contributing toward IoT and AI-based smart agriculture solutions for effective, sustainable corn field protection.","长尾猕猴（Macaca fascicularis）的侵害使玉米产量从潜在的每生长季7–9吨降至仅4–5吨（±40%），而诸如大声喧哗和射击等传统驱赶方法已被证明无效。本研究开发了一种基于物联网（IoT）的驱虫系统，采用声波、HC-SR501传感器和ESP32-CAM模块，并引入YOLOv8和EfficientNet-B0以替代先前使用的YOLOv3 Tiny模型，旨在提高检测精度并区分长尾猕猴与人类。由于合作农田的种群密度不足以构建足够的数据集，图像数据采集于南加里曼丹的Kembang岛，该岛作为自然栖息地，拥有大量且可观察的长尾猕猴种群，从而能够获取具有视觉代表性和多样性的训练数据。使用Label Studio对共计3,083张图像进行标注，分为两个类别（人类和长尾猕猴），然后按80:10:10的比例划分，未进行数据增强。YOLOv8n模型训练了50个epoch，之后将验证检测结果裁剪为920张图像，用于对EfficientNet-B0进行10个epoch的微调。评估结果表明，YOLOv8达到了0.959的精确率、0.965的召回率、0.979的mAP50和0.723的mAP50-95，而EfficientNet-B0达到了98.4–100%的验证准确率，且无明显过拟合。该两阶段检测-分类流水线在121张实地图像上成功进行了端到端测试，展示出相较于先前模型的显著性能提升，为基于物联网和人工智能的智慧农业解决方案做出了贡献，可有效、可持续地保护玉米田。","BEST Journal of Applied Electrical Science & Technology","2026-10-09T00:00:00Z",{"impact":16,"substance":278,"depth":18,"authority":20,"freshness":20,"relevant":21,"comment":393},"论文提出YOLOv8+EfficientNet-B0两阶段猕猴识别与声波驱避系统，数据规模与指标扎实，对智慧农业AI应用有参考价值。",[395],{"name":390,"url":387},[26,27,397,28,398],"物联网","玉米病虫害",{"code":95,"slug":400,"name":401,"path":402,"parent_code":8},"digital-agriculture","数字农业与前沿技术","\u002Ftopic\u002Fdigital-agriculture",[],{"status":405,"locked":13,"version":33,"method":34,"evidence":406},"partial",{"candidates":407,"agricultural_evidence":39,"reason":431},[408,412,419,424],{"code":38,"score":12,"supported":39,"title":409,"abstract":410,"tags":411},[41],[],[27,28],{"code":413,"score":12,"supported":39,"title":414,"abstract":415,"tags":418},"I5",[],[397,416,417],"internet of things","IoT",[397],{"code":45,"score":164,"supported":39,"title":420,"abstract":421,"tags":423},[],[26,422],"smart agriculture",[26],{"code":425,"score":67,"supported":39,"title":426,"abstract":427,"tags":430},"A3",[],[428,429],"玉米","corn",[],"一级领域明确，二级方向需要细分",[433,434],"长尾猕猴 玉米 识别","农业人工智能 玉米病虫害 智慧农业 目标检测","长尾猕猴玉米识别-4564","10.36456\u002F1hrtgc63",{"doi":436,"openalex_id":438,"authors":439,"venue":390,"cited_by_count":57,"oa_url":387,"card":447,"direction":451,"ingested_from":202},"W7221187629",[440,442,445],{"name":441,"orcid":8},"Hammami Rifqi Darmawan",{"name":443,"orcid":444},"Finki Dona Marleny","https:\u002F\u002Forcid.org\u002F0000-0002-3437-5594",{"name":446,"orcid":8},"Ayu Ahadi Ningrum",{"tldr":448,"method":449,"finding":450,"direction":451,"opportunity":452},"用YOLOv8+EfficientNet-B0两阶段流水线识别长尾猕猴，提升玉米田驱猴系统检测精度。","3083张标注图像，YOLOv8n检测后裁剪920张微调EfficientNet","YOLOv8 mAP50达0.979，EfficientNet-B0验证准确率98.4–100%，1","智慧农业 \u002F 农业物联网","可探索将两阶段检测-分类流水线轻量化部署到边缘设备，并扩展至多物种、多作物场景的实时驱避决策。","2026-10-10T23:30:07.974290Z"]