[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3667":3,"related-3667":56},{"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,"search_phrases":31,"slug":34,"view_count":35,"doi":36,"paper":37,"created_at":55},3667,"Real-Time Detection and Yield Estimation of Rosa damascena Using YOLOv10-N: A Case Study in the Dades Oasis, Drâa-Tafilalet, Southeastern Morocco","https:\u002F\u002Fdoi.org\u002F10.22194\u002Fjgias\u002F27.1790","Rosa damascena is an important economic resource in the Drâa Tafilalet region of southeast Morocco, particularly within the Dades Oasis in Kelaat M’Gouna. Detection and yield estimation of Rosa damascena flowers are important for increasing crop productivity and enhancing the local economy. In this work, we explored the application of YOLOv10-N, a deep learning model, for the real-time detection of Rosa damascena under varied environmental conditions. Utilizing a dataset of 2,807 annotated images captured under diverse lighting and terrain conditions, YOLOv10-N achieved an accuracy of 92.12%, recall of 77.22%, mAP50 of 90.65%, and mAP50-95 of 63.93%, demonstrating its robustness across varied environmental settings. In this study, we also present a novel method for estimating yield based on detected flower counts and average weight, providing a practical tool for better monitoring of rose fields, reducing labor costs, and optimizing resource management. These results demonstrate the importance of precision agriculture in addressing broader difficulties in agricultural productivity and economic sustainability. Keywords: YOLOv10, Rosa damascena, object detection, yield estimation, deep learning, computer vision in agriculture.","大马士革玫瑰（Rosa damascena）是摩洛哥东南部德拉-塔菲拉勒特地区的重要经济资源，尤其是在凯拉特姆古纳的达德斯绿洲。大马士革玫瑰花朵的检测与产量估测对于提高作物生产力和促进当地经济具有重要意义。在本研究中，我们探索了深度学习模型YOLOv10-N在不同环境条件下对大马士革玫瑰进行实时检测的应用。利用在多样光照和地形条件下采集的2,807张标注图像数据集，YOLOv10-N达到了92.12%的准确率、77.22%的召回率、90.65%的mAP50以及63.93%的mAP50-95，表明其在不同环境设置下具有稳健性。在本研究中，我们还提出了一种基于检测到的花朵数量和平均重量来估测产量的新方法，为更好地监测玫瑰田、降低劳动力成本和优化资源管理提供了实用工具。这些结果证明了精准农业在应对农业生产力和经济可持续性方面更广泛难题中的重要性。关键词：YOLOv10，大马士革玫瑰，目标检测，产量估测，深度学习，农业计算机视觉。",null,"Journal of Global Innovations in Agricultural Sciences","2026-09-25T00:00:00Z","论文",10,false,68,{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":20,"relevant":21,"comment":22},12,20,16,8,1,"基于YOLOv10-N的玫瑰实时检测与产量估算研究，数据集规模可观、指标完整，但属区域性案例，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","目标检测","产量估算","大马士革玫瑰",[32,33],"YOLOv10 大马士革玫瑰 产量估算","摩洛哥 Dades 绿洲 玫瑰检测","YOLOv10大马士革玫瑰产量估算-3667",0,"10.22194\u002Fjgias\u002F27.1790",{"doi":36,"openalex_id":38,"authors":39,"venue":10,"cited_by_count":35,"oa_url":46,"card":47,"direction":53,"ingested_from":54},"W7214280976",[40,43],{"name":41,"orcid":42},"Mohamed Ohamouddou","https:\u002F\u002Forcid.org\u002F0009-0000-9369-2891",{"name":44,"orcid":45},"Said Ohamouddou","https:\u002F\u002Forcid.org\u002F0009-0006-9175-9769","https:\u002F\u002Fjgiass.com\u002Fpdf-reader.php?file=Real-Time-Detection-and-Yield-Estimation-of-Rosa-damascena-Using-YOLOv10-N%3A-A-Case-Study-in-the-Dades-Oasis%2C-Dr%C3%A2a-Tafilalet%2C-Southeastern-Morocco.pdf&path=issue_papers",{"tldr":48,"method":49,"finding":50,"direction":51,"opportunity":52},"用YOLOv10-N实时检测大马士革玫瑰并基于花数估算产量。","2807张多光照地形标注图像训练YOLOv10-N，结合花数与均重估产。","检测精度92.12%、mAP50为90.65%，可跨环境稳健估产。","农业人工智能与决策模型","可扩展到多品种、多生长阶段及无人机边缘部署的实时估产研究。","农业遥感与作物表型","openalex","2026-09-28T23:30:20.312517Z",{"total":57,"page":21,"page_size":57,"items":58},6,[59,100,146,175,217,264],{"id":60,"title":61,"url":62,"summary":63,"summary_zh":64,"content":9,"source_name":65,"source_url":62,"published_at":66,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":67,"score_detail":68,"sources":75,"tags":77,"search_phrases":79,"slug":82,"view_count":21,"doi":83,"paper":84,"created_at":99},1382,"Instance Segmentation and Multi-Object Tracking for Fruit Quality Grading and Dynamic Yield Estimation in Egyptian Citrus Orchards","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102529","Egypt ranks among the world’s leading citrus producers and exporters; however, orchard operations remain constrained by subjective, labor-intensive manual quality grading and time-consuming yield estimation that require specialized expertise. This study presents an AI-driven orchard-intelligence framework for automated fruit quality grading and dynamic yield estimation under the unstructured conditions of Egyptian orange farms. To address the lack of in-field public datasets, a domain-specific instance-segmentation dataset of 618 original pixel-level annotated images, expanded to 1,545 augmented training images, and 15 videos was constructed. This dataset was used to benchmark single-stage segmentation models (YOLOv8, YOLOv11) against a two-stage (Mask R-CNN), alongside a transformer-based bounding-box detection (RT-DETR), under dense canopy occlusion, clustered fruit instances, and variable natural illumination. Each fine-tuned model was evaluated across two stages: static quality grading and dynamic yield estimation. For static grading (Good vs. Defective), YOLOv11l ranked among the top tier of evaluated segmenters, reaching mean Average Precision (mAP) of 0.932 with balanced precision-recall behavior, while maintaining real-time feasibility at 26.1 ms per image. For dynamic yield estimation, a multi-object tracking pipeline was implemented using BoT-SORT with camera-motion compensation and a virtual line-crossing counting mechanism. The evaluation reveals a three-way dissociation: the highest-mAP static segmenter (YOLOv11l), the best dynamic total counter (RT-DETR, WMAPE = 5.66%, despite the lowest box-level static accuracy), and the best quality-resolved dynamic grader (YOLOv11s, WMAPE = 6.97% at 14.2 ms per frame) are three different systems; RT-DETR did not resolve fruit quality reliably under motion. Counting accuracy and grading capability are thus separable properties requiring separate evaluation. Overall, the framework enables objective fruit quality assessment and actionable in-field fruit counting to support harvest logistics and labor planning, providing a practical foundation for smart agriculture in the Egyptian citrus sector.","埃及位居全球领先的柑橘生产国和出口国之列；然而，果园作业仍受制于主观性强、劳动密集的人工质量分级以及耗时且需专业技能的产量估算。本研究提出了一种人工智能驱动的果园智能框架，用于在埃及橙园的非结构化条件下实现自动化果实质量分级和动态产量估算。为解决田间公开数据集缺乏的问题，构建了一个领域特定的实例分割数据集，包含618张原始像素级标注图像，扩充至1545张增强训练图像，以及15段视频。该数据集用于在密集树冠遮挡、果实簇生和自然光照多变条件下，对单阶段分割模型（YOLOv8、YOLOv11）与两阶段模型（Mask R-CNN）进行基准测试，同时对比基于Transformer的边界框检测模型（RT-DETR）。每个微调模型均经过两个阶段的评估：静态质量分级和动态产量估算。在静态分级（优质与缺陷）中，YOLOv11l在评估的分割模型中位居前列，平均精度（mAP）达到0.932，精确率与召回率表现均衡，同时以每张图像26.1毫秒的处理速度保持实时可行性。在动态产量估算方面，采用基于BoT-SORT的多目标跟踪流程，结合相机运动补偿和虚拟线跨越计数机制。评估结果揭示了三重分离现象：静态分割精度最高的模型（YOLOv11l）、动态总数计数最优的模型（RT-DETR，加权平均绝对百分比误差WMAPE = 5.66%，尽管其框级静态精度最低），以及质量分辨动态分级最佳的模型（YOLOv11s，每帧14.2毫秒时WMAPE = 6.97%）是三个不同的系统；RT-DETR在运动条件下无法可靠分辨果实质量。因此，计数精度和分级能力是可分离的属性，需要分别评估。总体而言，该框架实现了客观的果实质量评估和可操作的田间果实计数，以支持采收物流和劳动力规划，为埃及柑橘产业的智慧农业提供了实用基础。","Smart Agricultural Technology","2026-09-01T00:00:00Z",79,{"impact":69,"substance":70,"depth":71,"authority":72,"freshness":73,"relevant":21,"comment":74},18,22,19,13,7,"针对埃及柑橘园提出AI驱动的品质分级与动态产量估算框架，方法新颖且数据详实，对智慧农业有参考价值。",[76],{"name":65,"url":62},[26,27,28,78,29],"柑橘",[80,81],"农业人工智能 产量估算 智慧农业 目标检测","农业人工智能 产量估算","农业人工智能产量估算智慧农业目标检测-1382","10.1016\u002Fj.atech.2026.102529",{"doi":83,"openalex_id":85,"authors":86,"venue":65,"cited_by_count":35,"oa_url":62,"card":93,"direction":97,"ingested_from":54},"W7204959516",[87,89,91],{"name":88,"orcid":9},"Saleh A. Farhoud",{"name":90,"orcid":9},"Yasmine A. Zaghloul",{"name":92,"orcid":9},"Omar M. Shehata",{"tldr":94,"method":95,"finding":96,"direction":97,"opportunity":98},"构建柑橘果园AI框架，实现果实质量分级与动态产量估计，并评估多种分割与跟踪模型。","构建实例分割数据集，微调YOLOv8\u002Fv11、Mask R-CNN、RT-DET","静态分级最优为YOLOv11l，动态计数最优为RT-DETR，质量动态分级最优为YOLOv11s，计","智慧农业 \u002F 农业物联网","可探索多任务协同优化，如设计统一模型同时满足静态分级与动态计数，或引入时序信息提升运动场景下的质量分级鲁棒性。","2026-09-02T23:30:04.506009Z",{"id":101,"title":102,"url":103,"summary":104,"summary_zh":105,"content":9,"source_name":106,"source_url":103,"published_at":107,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":108,"score_detail":109,"sources":113,"tags":115,"search_phrases":118,"slug":121,"view_count":35,"doi":122,"paper":123,"created_at":145},3356,"Simultaneous Maturity Recognition and Yield Counting of Truss and Individual Tomatoes Using Improved YOLOv8-EME and Optimized ByteTrack Tracker","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fhorticulturae12101199","Crop maturity detection and yield estimation are critical components of protected agriculture, supporting optimized harvest timing, fruit quality control, and coordination of production and marketing. Existing studies on detection and counting predominantly address a single category—either truss or individual fruits. To advance automation and intelligence in production management, this study proposes a tomato detection and counting system that integrates an improved YOLOv8-EME model with the ByteTrack algorithm, enabling simultaneous detection and counting of truss and individual fruits with maturity classification. The improved YOLOv8-EME model combines the EfficientNet architecture with the EffectiveSE attention mechanism, improving feature extraction and computational efficiency. In addition, the optimized network structure yields a lightweight design, reducing FLOPs to 6.9 G. The model attains a mean Average Precision (mAP) of 0.942, 0.883, 0.850, and 0.956 for the Ripe, Raw, Medium-Raw, and Truss categories, respectively. A proposed cross-line counting method integrated with an improved ByteTrack algorithm mitigates target loss, ID drift, and duplicate counting through ID drift association, trajectory fusion, historical trajectory cues, and a cooldown scheme. These designs significantly improve detection and counting accuracy for truss and fruit maturity, achieving a counting accuracy of 94%. The system offers efficient and accurate technical support for tomato detection and counting in smart agriculture.","作物成熟度检测与产量估测是设施农业的关键环节，可为优化采收时机、果实品质控制及产销协调提供支撑。现有检测与计数研究大多仅针对单一类别，即串收番茄或单个果实。为推动生产管理的自动化与智能化，本研究提出了一种融合改进YOLOv8-EME模型与ByteTrack算法的番茄检测与计数系统，可实现串收番茄与单个果实的同步检测与计数，并进行成熟度分类。改进后的YOLOv8-EME模型将EfficientNet架构与EffectiveSE注意力机制相结合，提升了特征提取能力与计算效率。此外，优化后的网络结构实现了轻量化设计，浮点运算次数（FLOPs）降至6.9 G。该模型在成熟、未成熟、半熟和串收四个类别上的平均精度均值（mAP）分别为0.942、0.883、0.850和0.956。所提出的跨线计数方法与改进的ByteTrack算法相结合，通过ID漂移关联、轨迹融合、历史轨迹线索和冷却机制，缓解了目标丢失、ID漂移和重复计数问题。这些设计显著提升了串收番茄和果实成熟度的检测与计数精度，计数准确率达到94%。该系统为智慧农业中的番茄检测与计数提供了高效、准确的技术支持。","Horticulturae","2026-09-23T00:00:00Z",77,{"impact":110,"substance":70,"depth":69,"authority":72,"freshness":111,"relevant":21,"comment":112},15,9,"该研究提出改进YOLOv8-EME与优化ByteTrack的番茄检测计数系统，可同时识别串收与单果成熟度并计数，方法新颖、数据详实，对智慧农业采摘自动化有实用价值。",[114],{"name":106,"url":103},[26,27,28,116,117],"番茄","产量估测",[119,120],"YOLOv8 番茄 成熟度检测","番茄 串收 产量计数","YOLOv8番茄成熟度检测-3356","10.3390\u002Fhorticulturae12101199",{"doi":122,"openalex_id":124,"authors":125,"venue":106,"cited_by_count":35,"oa_url":103,"card":140,"direction":97,"ingested_from":54},"W7214123905",[126,128,131,133,135,137],{"name":127,"orcid":9},"Liying Shi",{"name":129,"orcid":130},"Sen Lin","https:\u002F\u002Forcid.org\u002F0000-0001-6521-3152",{"name":132,"orcid":9},"Haihang Zhao",{"name":134,"orcid":9},"Tianlong Sun",{"name":136,"orcid":9},"Dongdong Sun",{"name":138,"orcid":139},"Yuchen Yang","https:\u002F\u002Forcid.org\u002F0000-0001-5977-1617",{"tldr":141,"method":142,"finding":143,"direction":51,"opportunity":144},"提出改进YOLOv8-EME与ByteTrack结合的番茄检测计数系统，可同时识别串收与单果成熟度并","改进YOLOv8-EME（EfficientNet+EffectiveSE）结合","模型mAP达0.942\u002F0.883\u002F0.850\u002F0.956，计数准确率94%，FLOPs仅6.9G。","可探索多作物、多生长阶段通用模型，并融合边缘部署与产量预测决策。","2026-09-24T23:30:10.615639Z",{"id":147,"title":148,"url":149,"summary":150,"summary_zh":9,"content":9,"source_name":151,"source_url":9,"published_at":152,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":153,"score_detail":154,"sources":157,"tags":159,"search_phrases":162,"slug":165,"view_count":35,"doi":9,"paper":166,"created_at":174},3322,"PB-DETR：群养生猪多行为识别的自适应液体神经网络编码器与动态监督框架","https:\u002F\u002Fjky.yangtzeu.edu.cn\u002Finfo\u002F1265\u002F6884.htm","长江大学计算机科学学院研究生郭宇衡作为第一作者，在计算机与农业交叉领域顶级期刊《Computers and Electronics in Agriculture》发表题为PB-DETR: multi-behaviour recognition of group-housed pigs with adaptive liquid neural network encoder and dynamic supervision的研究论文。长江大学智慧农业交叉创新研究中心主任、计科学院詹炜教授为通讯作者。研究将液体神经网络机制引入Transformer编码器，设计自适应液体时间常数增强模块，构建多尺度特征交互和动态质量感知优化策略，进而降低模型对固定背景信息的依赖。基于PB-DETR框架构建的模型平均精度(mAP)达到0.893，参数量为14.58M，计算量为49.12GFLOPs；在不同养殖场景之间的测试中表现出较好稳定性。","《Computers and Electronics in Agriculture》2026","2026-09-21T00:00:00Z",78,{"impact":19,"substance":70,"depth":69,"authority":155,"freshness":20,"relevant":21,"comment":156},14,"农业人工智能顶刊论文，方法新颖、指标明确，对智慧养殖行为识别有实质参考价值，值得进入每日精选。",[158],{"name":151,"url":149},[26,27,160,28,161],"智慧养殖","生猪行为识别",[163,164],"长江大学 詹炜 生猪行为识别","PB-DETR 群养生猪 多行为识别","长江大学詹炜生猪行为识别-3322",{"doi":9,"openalex_id":9,"authors":167,"venue":9,"cited_by_count":35,"oa_url":9,"card":168,"direction":51,"ingested_from":173},[],{"tldr":169,"method":170,"finding":171,"direction":51,"opportunity":172},"提出PB-DETR框架，实现群养生猪多行为识别，mAP达0.893。","引入液体神经网络改进Transformer编码器，结合多尺度特征交互与动态监督。","模型在群养生猪多行为识别中mAP达0.893，跨场景稳定性较好。","可探索液体神经网络在其它畜禽行为识别中的泛化能力及轻量化部署。","agent","2026-09-24T00:04:02.413623Z",{"id":176,"title":177,"url":178,"summary":179,"summary_zh":180,"content":9,"source_name":181,"source_url":178,"published_at":182,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":183,"score_detail":184,"sources":186,"tags":188,"search_phrases":191,"slug":194,"view_count":35,"doi":195,"paper":196,"created_at":216},2798,"Deep Learning-Based Small-Object Detection in UAV Imagery: A Symmetry-Informed Review of Scale Variation and Scenario Constraints","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fsym18091548","As UAV platforms are increasingly employed in military reconnaissance, disaster monitoring, precision agriculture, and other remote-sensing applications, small-object detection in UAV imagery is a critical yet challenging task. Existing reviews primarily address either general small-object detection or general object detection in UAV imagery, whereas reviews specifically devoted to small-object detection in UAV imagery remain scarce. This review examines existing deep learning-based research on small-object detection in UAV imagery from a problem-driven and deployment-oriented perspective. A symmetry-informed perspective was also adopted to analyze how variations in object-scale, viewpoint, imaging quality, and scene conditions affected feature representation and detection stability. Building on five major challenges in UAV small-object detection, a three-tier framework encompassing detection paradigms, key technical strategies, and scenario-specific constraints was established. Representative UAV datasets were further examined to characterize object-scale distributions, while existing methods were reviewed in terms of multiscale representation, feature enhancement, training optimization, and lightweight deployment. The applicability of different technical approaches was then discussed across five typical UAV scenarios. In addition, under relatively consistent experimental conditions, representative detection models based on two-stage, one-stage, and Transformer-based paradigms were analyzed in terms of detection accuracy, computational efficiency, and deployment feasibility. Finally, current limitations and future directions were summarized for scale-robust, scenario-adaptive, and resource-aware small-object detection in UAV imagery.","随着无人机平台在军事侦察、灾害监测、精准农业等遥感应用中的日益普及，无人机图像中的小目标检测成为一项关键但具有挑战性的任务。现有综述主要关注通用小目标检测或无人机图像中的通用目标检测，而专门针对无人机图像中小目标检测的综述仍然匮乏。本综述从问题驱动和部署导向的视角，审视了现有的基于深度学习的无人机图像小目标检测研究。同时采用对称性视角，分析了目标尺度、视角、成像质量和场景条件的变化如何影响特征表示和检测稳定性。基于无人机小目标检测中的五大挑战，构建了一个涵盖检测范式、关键技术策略和场景特定约束的三层框架。进一步考察了代表性无人机数据集以刻画目标尺度分布，同时从多尺度表示、特征增强、训练优化和轻量化部署等方面综述了现有方法。随后讨论了不同技术方法在五种典型无人机场景中的适用性。此外，在相对一致的实验条件下，对基于两阶段、单阶段和Transformer范式的代表性检测模型在检测精度、计算效率和部署可行性方面进行了分析。最后，总结了当前局限性及未来方向，以推动无人机图像中尺度鲁棒、场景自适应和资源感知的小目标检测研究。","Symmetry","2026-09-16T00:00:00Z",76,{"impact":19,"substance":18,"depth":69,"authority":72,"freshness":111,"relevant":21,"comment":185},"无人机小目标检测的系统性综述，提出对称性视角与三层框架，对农业遥感与精准农业部署有实质参考价值。",[187],{"name":181,"url":178},[26,189,27,190,28],"无人机","农业遥感",[192,193],"农业人工智能 农业遥感 智慧农业 目标检测","农业人工智能 农业遥感","农业人工智能农业遥感智慧农业目标检测-2798","10.3390\u002Fsym18091548",{"doi":195,"openalex_id":197,"authors":198,"venue":181,"cited_by_count":35,"oa_url":178,"card":211,"direction":53,"ingested_from":54},"W7213430256",[199,201,204,206,208],{"name":200,"orcid":9},"Ling Wen",{"name":202,"orcid":203},"Anmin Gong","https:\u002F\u002Forcid.org\u002F0000-0002-2715-2296",{"name":205,"orcid":9},"Caihong Ma",{"name":207,"orcid":9},"Shengzhou Ma",{"name":209,"orcid":210},"Yanzhe Zhang","https:\u002F\u002Forcid.org\u002F0009-0008-6529-1247",{"tldr":212,"method":213,"finding":214,"direction":53,"opportunity":215},"综述无人机图像小目标检测的深度学习研究，提出对称性视角与三层框架。","文献综述，分析代表性数据集与两阶段、单阶段、Transformer模型。","尺度变化和场景约束是核心挑战，需尺度鲁棒、场景自适应和资源感知检测。","可探索农业场景下尺度自适应与轻量化部署的平衡，以及跨场景泛化的小目标检测。","2026-09-17T23:30:37.479759Z",{"id":218,"title":219,"url":220,"summary":221,"summary_zh":222,"content":9,"source_name":223,"source_url":220,"published_at":182,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":224,"sources":226,"tags":228,"search_phrases":230,"slug":233,"view_count":35,"doi":234,"paper":235,"created_at":263},2741,"A tomato maturity detection method against occlusion and variable illumination","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112423","A tomato maturity detection method against occlusion and variable illumination。Computers and Electronics in Agriculture","一种抗遮挡和可变光照的番茄成熟度检测方法","Computers and Electronics in Agriculture",{"impact":17,"substance":69,"depth":19,"authority":155,"freshness":20,"relevant":21,"comment":225},"核心期刊论文，针对遮挡与光照变化下的番茄成熟度检测提出新方法，属农业人工智能细分领域实质进展，但应用范围有限，未达重大突破层级。",[227],{"name":223,"url":220},[26,27,28,116,229],"作物表型",[231,232],"农业人工智能 作物表型 智慧农业 目标检测","农业人工智能 作物表型","农业人工智能作物表型智慧农业目标检测-2741","10.1016\u002Fj.compag.2026.112423",{"doi":234,"openalex_id":236,"authors":237,"venue":223,"cited_by_count":35,"oa_url":9,"card":258,"direction":51,"ingested_from":54},"W7213429689",[238,241,244,247,250,253,255],{"name":239,"orcid":240},"Hao Meng","https:\u002F\u002Forcid.org\u002F0000-0001-7511-2910",{"name":242,"orcid":243},"Wenzhe Li","https:\u002F\u002Forcid.org\u002F0009-0008-1630-4697",{"name":245,"orcid":246},"Di Wang","https:\u002F\u002Forcid.org\u002F0000-0002-3911-8159",{"name":248,"orcid":249},"Hui Zhao","https:\u002F\u002Forcid.org\u002F0009-0005-4192-5776",{"name":251,"orcid":252},"Ximing Li","https:\u002F\u002Forcid.org\u002F0000-0003-4022-1273",{"name":254,"orcid":9},"Dongdong Cui",{"name":256,"orcid":257},"Fernando Auat Cheein","https:\u002F\u002Forcid.org\u002F0000-0002-6347-7696",{"tldr":259,"method":260,"finding":261,"direction":51,"opportunity":262},"提出一种抗遮挡和光照变化的番茄成熟度检测方法。","基于深度学习的图像检测，针对遮挡与光照变化优化。","该方法在遮挡和变光照下仍能准确检测番茄成熟度。","可探索多模态融合与轻量化部署，提升田间复杂场景实时检测鲁棒性。","2026-09-17T23:30:01.491379Z",{"id":265,"title":266,"url":267,"summary":268,"summary_zh":269,"content":9,"source_name":270,"source_url":267,"published_at":182,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":183,"score_detail":271,"sources":275,"tags":277,"search_phrases":280,"slug":283,"view_count":21,"doi":284,"paper":285,"created_at":303},2638,"Full shuffle and p-rectified semi-inner powerful IoU-based chili pepper flower recognition with environment-aware YOLO11","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1935123","As a common greenhouse-grown commercial crop, chili pepper yields have always been a focus of attention. Effective management is essential for both environmental sustainability and high productivity in greenhouses. Whether for growth monitoring, yield estimates, and automated production, precise identification of chili flowers is critical. To this goal, this paper works upon the YOLO11 object detection model, focusing on the precise detection of chili flowers. The diversity of samples across various scenarios is increased by utilizing a self-built dataset of chili flowers in greenhouses and integrating data augmentation. To address YOLO11’s limitations in this task, this paper proposes a unified framework that integrates three modules: (1) full shuffle to enhance information exchange between model channels and optimize weights; (2) Since YOLO11 cannot incorporate external environmental data during training, the environment-aware C3k2 is introduced; and (3) The proposed p-rectified Semi-inner Powerful IoU accelerates model convergence and flexibility. Experiments show that our method obtains 84.8% mAP50, 50.2% mAP50-95, 81.1% Precision, and 75.6% Recall, outperforming the baseline YOLO11’s 77.3% by 7.5 percentage points, as well as YOLOv8, YOLO12, RTDETR, and Faster R-CNN under the same experimental conditions. Therefore, in terms of performance, our method generally surpasses existing algorithms for chili flower detection. Research on detecting chili flowers in greenhouses remains limited. The study offers valuable insights for the advancement of smart agriculture. Future work will further explore the model’s generalization capabilities across multiple varieties and growth stages, as well as deploy it on embedded devices for practical application.","作为常见的温室商业化种植作物，辣椒的产量一直备受关注。有效的管理对于温室的環境可持续性和高生产力都至关重要。无论是生长监测、产量估算还是自动化生产，辣椒花朵的精准识别都至关重要。为实现这一目标，本文基于YOLO11目标检测模型，聚焦辣椒花朵的精准检测。通过利用自建的温室辣椒花朵数据集并结合数据增强，增加了不同场景下样本的多样性。针对YOLO11在此任务中的局限性，本文提出了一个集成三个模块的统一框架：（1）完全混洗（full shuffle），以增强模型通道间的信息交换并优化权重；（2）由于YOLO11在训练过程中无法纳入外部环境数据，引入了环境感知C3k2（environment-aware C3k2）；（3）提出的p-rectified Semi-inner Powerful IoU加速了模型收敛并提升了灵活性。实验表明，本方法取得了84.8%的mAP50、50.2%的mAP50-95、81.1%的精确率和75.6%的召回率，较基线YOLO11的77.3%提升了7.5个百分点，并在相同实验条件下优于YOLOv8、YOLO12、RTDETR和Faster R-CNN。因此，在性能方面，本方法总体上超越了现有的辣椒花朵检测算法。目前关于温室辣椒花朵检测的研究仍然有限。本研究为智慧农业的发展提供了有价值的见解。未来工作将进一步探索模型在多个品种和生长阶段上的泛化能力，并将其部署到嵌入式设备上以实现实际应用。","Frontiers in Plant Science",{"impact":110,"substance":272,"depth":273,"authority":72,"freshness":13,"relevant":21,"comment":274},21,17,"基于YOLO11的辣椒花检测新方法，mAP50提升7.5个百分点，对设施农业智能监测有参考价值，但属细分技术进展，影响范围有限。",[276],{"name":270,"url":267},[26,27,278,28,279],"设施农业","辣椒",[281,282],"农业人工智能 智慧农业 目标检测 设施农业","农业人工智能 智慧农业","农业人工智能智慧农业目标检测设施农业-2638","10.3389\u002Ffpls.2026.1935123",{"doi":284,"openalex_id":286,"authors":287,"venue":270,"cited_by_count":35,"oa_url":267,"card":298,"direction":97,"ingested_from":54},"W7213351973",[288,290,292,295],{"name":289,"orcid":9},"Cui-Ping Zhang",{"name":291,"orcid":9},"Zhi-Yong Wang",{"name":293,"orcid":294},"Xuewei Wang","https:\u002F\u002Forcid.org\u002F0000-0001-9604-3045",{"name":296,"orcid":297},"Zhi Li","https:\u002F\u002Forcid.org\u002F0000-0001-5571-0518",{"tldr":299,"method":300,"finding":301,"direction":51,"opportunity":302},"基于YOLO11改进，实现温室辣椒花精准检测。","自建数据集+数据增强，引入全混洗、环境感知C3k2和p校正IoU。","mAP50达84.8%，比基线YOLO11提升7.5个百分点，优于多个对比模型。","温室辣椒花检测研究少，可探索多品种、多生长期泛化及嵌入式部署。","2026-09-16T23:30:09.868848Z"]