[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3165":3,"related-3165":127},{"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":24,"tags":26,"primary_topic":32,"secondary_topics":38,"classification":45,"search_phrases":84,"slug":87,"view_count":88,"doi":89,"paper":90,"created_at":126},3165,"Explainable growth stage classification of cacao (Theobroma cacao L.) leaves and key feature visualization using vision transformer and transfer learning","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1885906","Accurate and objective plant phenotyping is crucial for optimizing agricultural practices, understanding plant development, and enabling rapid responses to environmental changes. Traditional methods, often relying on visual observation, can be subjective, time-consuming, and may overlook subtle but important differences. This study demonstrates the power of combining digital imaging with deep learning to classify plant material with high accuracy, even when visual differences are minimal. We focused on differentiating between stage D (light green) and stage E (dark green) leaves of cacao ( Theobroma cacao L. ), which are visually very similar in size and overall structure. Using cleared and stained leaves of the SCA 6 genotype to highlight the venation network, we trained a Vision Transformer (ViT) model, a deep learning architecture, on image patches. At the patch level, the model achieved an overall accuracy of 97.03% on an independent test set, with a recall of 96.0% for stage D and 98.3% for stage E. At the whole-leaf level, majority voting correctly classified 14 of 15 independent test leaves (93.3%). Attention maps indicated that image regions containing the midrib and primary lateral veins contributed strongly to classification. These attention maps identify discriminative image regions, but they do not by themselves determine the biological mechanism underlying the signal. The major-vein signal may reflect developmental differences in vein-associated structure, stage-associated differences in Safranin O uptake or optical density, tissue thickness, or a combination of these factors. Because vascular anatomy, lignification, hydraulic conductance, phloem loading, and source–sink status were not directly measured, these mechanisms are treated as hypotheses requiring future anatomical, histochemical, and physiological validation. Thus, this study provides a proof-of-concept for interpretable image-based classification of stage D and stage E leaves within greenhouse-grown SCA 6 cacao. Extension to other cacao genotypes, field-grown plants, independent seasons, staining batches, stress detection, species identification, genotype discrimination, or precision-agriculture deployment will require external validation.","准确、客观的植物表型分析对于优化农业实践、理解植物发育以及快速响应环境变化至关重要。传统方法通常依赖视觉观察，可能具有主观性、耗时，并且可能忽略细微但重要的差异。本研究展示了将数字成像与深度学习相结合，即使在视觉差异极小的情况下，也能以高精度对植物材料进行分类。我们聚焦于区分可可（Theobroma cacao L.）的D期（浅绿色）和E期（深绿色）叶片，这些叶片在大小和整体结构上视觉上非常相似。利用SCA 6基因型的透明染色叶片以突出脉序网络，我们在图像块上训练了Vision Transformer（ViT）模型，一种深度学习架构。在图像块水平上，该模型在独立测试集上达到了97.03%的总体准确率，D期的召回率为96.0%，E期为98.3%。在整叶水平上，多数投票正确分类了15片独立测试叶片中的14片（93.3%）。注意力图表明，包含中脉和初级侧脉的图像区域对分类贡献显著。这些注意力图识别了具有判别力的图像区域，但它们本身并不能确定信号背后的生物学机制。主脉信号可能反映了脉相关结构的发育差异、番红O摄取或光密度的阶段相关差异、组织厚度，或这些因素的组合。由于未直接测量维管解剖结构、木质化、水力导度、韧皮部装载和源–库状态，这些机制被视为假设，需要未来的解剖学、组织化学和生理学验证。因此，本研究为温室种植的SCA 6可可中D期和E期叶片的可解释图像分类提供了概念验证。扩展到其他可可基因型、田间种植植株、独立季节、染色批次、胁迫检测、物种鉴定、基因型区分或精准农业部署将需要外部验证。",null,"Frontiers in Plant Science","2026-09-21T00:00:00Z","论文",10,false,71,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},12,20,17,13,9,1,"方法新颖、数据可靠的可解释作物表型概念验证研究，但属实验室小样本，产业影响有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","深度学习","可可","作物表型",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"B3","tea-coffee","茶叶与咖啡","\u002Ftopic\u002Fhorticulture\u002Ftea-coffee","B",[39],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":46,"locked":14,"version":47,"method":48,"evidence":49},"classified","2026-10-22.2","rules",{"candidates":50,"agricultural_evidence":53},[51,58,64,72,79],{"code":33,"score":52,"supported":53,"title":54,"abstract":56,"tags":57},15,true,[55],"cacao",[30,55],[30],{"code":40,"score":59,"supported":53,"title":60,"abstract":61,"tags":63},8,[],[29,62],"deep learning",[28,29],{"code":65,"score":66,"supported":53,"title":67,"abstract":68,"tags":71},"B7",6,[],[69,70],"温室","greenhouse",[],{"code":73,"score":74,"supported":53,"title":75,"abstract":76,"tags":78},"I1",4,[],[77],"精准农业",[27],{"code":80,"score":22,"supported":14,"title":81,"abstract":82,"tags":83},"I3",[],[],[31],[85,86],"可可 叶片 生长阶段 分类","Vision Transformer 作物表型","可可叶片生长阶段分类-3165",5,"10.3389\u002Ffpls.2026.1885906",{"doi":89,"openalex_id":91,"authors":92,"venue":10,"cited_by_count":118,"oa_url":6,"card":119,"direction":123,"ingested_from":125},"W7213901640",[93,95,98,100,102,104,107,110,113,115],{"name":94,"orcid":9},"Ezekiel Ahn",{"name":96,"orcid":97},"Eun-Sung Park","https:\u002F\u002Forcid.org\u002F0000-0001-6826-2865",{"name":99,"orcid":9},"Moon S. Kim",{"name":101,"orcid":9},"Hangi Kim",{"name":103,"orcid":9},"Lalit M. Kandpal",{"name":105,"orcid":106},"Sunchung Park","https:\u002F\u002Forcid.org\u002F0000-0002-7398-9476",{"name":108,"orcid":109},"Seunghyun Lim","https:\u002F\u002Forcid.org\u002F0000-0003-3023-4863",{"name":111,"orcid":112},"Lyndel W. Meinhardt","https:\u002F\u002Forcid.org\u002F0000-0001-8299-2629",{"name":114,"orcid":9},"Byoung-Kwan Cho",{"name":116,"orcid":117},"Insuck Baek","https:\u002F\u002Forcid.org\u002F0000-0003-1044-349X",0,{"tldr":120,"method":121,"finding":122,"direction":123,"opportunity":124},"用ViT和迁移学习对可可叶D、E期进行可解释分类，准确率达97%。","透明染色叶片图像块训练ViT，注意力图可视化关键区域。","模型准确区分D\u002FE期，中脉和主侧脉区域贡献最大。","农业遥感与作物表型","可扩展到多基因型、田间、胁迫检测，并验证脉信号生物学机制。","openalex","2026-09-22T23:30:19.972949Z",{"total":66,"page":22,"page_size":66,"items":128},[129,219,275,338,393,477],{"id":130,"title":131,"url":132,"summary":133,"summary_zh":134,"content":9,"source_name":135,"source_url":132,"published_at":136,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":137,"score_detail":138,"sources":143,"tags":145,"primary_topic":149,"secondary_topics":150,"classification":152,"search_phrases":183,"slug":186,"view_count":22,"doi":187,"paper":188,"created_at":218},4081,"Aerial Phenotyping of Coffee Flowering Using Deep Learning for Early Yield Prediction","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fplants15193019","Early yield estimation is important for improving decision-making and management practices in coffee production systems. However, approaches that use aerial imagery to characterize flowering and relate it to subsequent yield remain limited. This study evaluated the potential of RGB imagery acquired by remotely piloted aircraft (RPA) combined with Deep Learning techniques to classify coffee plants according to three levels of a flowering proxy and to investigate its relationship with yield. A dataset comprising 1593 segmented RGB images was classified into Low, Medium, and High flowering levels using a custom convolutional neural network and three transfer learning architectures: VGG16, ResNet50V2, and MobileNetV2. ResNet50V2 was selected for final evaluation based on its validation performance and achieved 92.12% accuracy, 92.40% macro-precision, 91.81% macro-recall, and 92.06% macro-F1 on the test dataset. The flowering proxy showed a positive association with field-observed yield, with Spearman’s ρ = 0.701 (p \u003C 0.001) and R2 = 0.528 for the combined dataset of Arara and Mundo Novo. In consolidated out-of-sample predictions, the model achieved R2 = 0.489, MAE = 1.03 kg plant−1, and RMSE = 1.18 kg plant−1. Performance differed between cultivars, with consolidated R2 values of 0.567 for Arara and 0.329 for Mundo Novo. Spatial analysis also showed correspondence between the distribution of flowering levels derived from RGB imagery and the subsequent spatial distribution of yield. These findings indicate that RPA-based RGB imagery and Deep Learning can provide spatialized information related to coffee flowering and its productive performance, with potential applications in precision agriculture. Further evaluation across production cycles, cultivars, and environmental conditions is needed to assess the robustness and generalization of the approach.","早期产量估测对于改善咖啡生产系统中的决策与管理实践具有重要意义。然而，利用航空影像表征开花并将其与后续产量相关联的方法仍然有限。本研究评估了由遥控驾驶航空器（RPA）获取的RGB影像结合深度学习技术，根据开花代理指标的三个水平对咖啡植株进行分类，并探讨其与产量之间关系的潜力。利用自定义卷积神经网络和三种迁移学习架构——VGG16、ResNet50V2和MobileNetV2，将包含1593张分割RGB图像的数据集分类为低、中、高开花水平。基于验证性能，选择ResNet50V2进行最终评估，其在测试数据集上达到了92.12%的准确率、92.40%的宏精确率、91.81%的宏召回率和92.06%的宏F1值。开花代理指标与田间观测产量呈正相关，Arara和Mundo Novo合并数据集的Spearman's ρ = 0.701（p \u003C 0.001），R² = 0.528。在合并的样本外预测中，模型达到了R² = 0.489、MAE = 1.03 kg plant⁻¹和RMSE = 1.18 kg plant⁻¹。不同品种间的表现存在差异，Arara和Mundo Novo的合并R²值分别为0.567和0.329。空间分析还表明，由RGB影像得出的开花水平分布与后续产量的空间分布之间存在对应关系。这些发现表明，基于RPA的RGB影像和深度学习能够提供与咖啡开花及其生产性能相关的空间化信息，在精准农业中具有应用潜力。需要在不同生产周期、品种和环境条件下进行进一步评估，以检验该方法的稳健性和泛化能力。","Plants","2026-10-02T00:00:00Z",78,{"impact":139,"substance":140,"depth":141,"authority":20,"freshness":21,"relevant":22,"comment":142},16,22,18,"该研究将无人机RGB影像与深度学习结合用于咖啡花期分类和早期产量预测，方法新颖、数据扎实、结论可靠，对智慧农业和精准农业有参考价值，但属于细分作物研究，影响范围有限。",[144],{"name":135,"url":132},[27,28,146,147,148],"产量预测","遥感","咖啡种植",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[151],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},{"status":46,"locked":14,"version":47,"method":48,"evidence":153},{"candidates":154,"agricultural_evidence":53},[155,161,165,171,178],{"code":33,"score":52,"supported":53,"title":156,"abstract":158,"tags":160},[157],"coffee",[159,157],"咖啡",[148],{"code":40,"score":52,"supported":53,"title":162,"abstract":163,"tags":164},[62],[29,62],[28],{"code":73,"score":166,"supported":53,"title":167,"abstract":168,"tags":170},7,[],[77,169],"precision agriculture",[27],{"code":172,"score":66,"supported":53,"title":173,"abstract":174,"tags":177},"I4",[],[175,176],"空间分析","spatial analysis",[],{"code":80,"score":179,"supported":14,"title":180,"abstract":181,"tags":182},2,[],[],[146,147],[184,185],"无人机 RGB 咖啡 花期 产量预测","ResNet50V2 咖啡 深度学习","无人机RGB咖啡花期产量预测-4081","10.3390\u002Fplants15193019",{"doi":187,"openalex_id":189,"authors":190,"venue":135,"cited_by_count":118,"oa_url":132,"card":213,"direction":123,"ingested_from":125},"W7216244707",[191,193,196,199,201,204,207,210],{"name":192,"orcid":9},"Elias Horácio Zavala",{"name":194,"orcid":195},"Gabriel Araújo e Silva Ferraz","https:\u002F\u002Forcid.org\u002F0000-0001-6403-2210",{"name":197,"orcid":198},"Felipe Andrés Obando Vega","https:\u002F\u002Forcid.org\u002F0000-0001-7241-7065",{"name":200,"orcid":9},"Mirian de Lourdes Oliveira",{"name":202,"orcid":203},"Milene Alves de Figueiredo Carvalho","https:\u002F\u002Forcid.org\u002F0000-0001-6138-8182",{"name":205,"orcid":206},"Samuel Henrique Braga da Cunha","https:\u002F\u002Forcid.org\u002F0000-0002-8783-0378",{"name":208,"orcid":209},"Andrea Pagliai","https:\u002F\u002Forcid.org\u002F0000-0002-0455-7426",{"name":211,"orcid":212},"Daniele Sarri","https:\u002F\u002Forcid.org\u002F0000-0001-6990-7573",{"tldr":214,"method":215,"finding":216,"direction":123,"opportunity":217},"用无人机RGB影像和深度学习对咖啡开花分级，实现早期产量预测。","1593张分割RGB图，自定义CNN与VGG16\u002FResNet50V2\u002FMobi","ResNet50V2开花分级准确率92.12%，开花等级与产量正相关（ρ=0.701）。","可跨生产周期、品种和环境验证模型泛化性，并融合多源遥感提升早期产量预测稳健性。","2026-10-03T23:30:27.930711Z",{"id":220,"title":221,"url":222,"summary":223,"summary_zh":9,"content":9,"source_name":224,"source_url":9,"published_at":225,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":226,"score_detail":227,"sources":229,"tags":231,"primary_topic":235,"secondary_topics":236,"classification":242,"search_phrases":262,"slug":265,"view_count":22,"doi":9,"paper":266,"created_at":274},1367,"基于无人机多源遥感特征的茶树生长与氮素状况深度学习监测","https:\u002F\u002Fwww.mdpi.com\u002F2072-4292\u002F18\u002F17\u002F2923","南京农业大学园艺学院团队提出融合光谱、纹理与谐波特征的多域融合指标，结合无人机数据与深度学习模型，用于茶树生物量与氮素积累量估算。实验表明多域融合指数显著优于传统光谱指标，可实现对茶园生物量与氮素状况的高精度反演，为近实时茶园监测提供工具。","Remote Sensing · MDPI · 2026-09-01","2026-08-31T16:00:00Z",72,{"impact":52,"substance":18,"depth":141,"authority":17,"freshness":166,"relevant":22,"comment":228},"研究提出多域融合指标，提升茶树氮素监测精度，对智慧茶园有实用价值。",[230],{"name":224,"url":222},[27,232,147,233,234],"无人机","氮素监测","茶树",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[237,238],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},{"code":80,"slug":239,"name":240,"path":241,"parent_code":44},"remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":46,"locked":14,"version":47,"method":48,"evidence":243},{"candidates":244,"agricultural_evidence":53},[245,250,254,258],{"code":33,"score":246,"supported":53,"title":247,"abstract":248,"tags":249},11,[234],[234],[],{"code":40,"score":246,"supported":53,"title":251,"abstract":252,"tags":253},[29],[29],[],{"code":80,"score":21,"supported":53,"title":255,"abstract":256,"tags":257},[147],[],[147],{"code":73,"score":22,"supported":14,"title":259,"abstract":260,"tags":261},[],[],[27],[263,264],"智慧农业 氮素监测 无人机 茶树","智慧农业 氮素监测","智慧农业氮素监测无人机茶树-1367",{"doi":9,"openalex_id":9,"authors":267,"venue":9,"cited_by_count":118,"oa_url":9,"card":268,"direction":123,"ingested_from":273},[],{"tldr":269,"method":270,"finding":271,"direction":123,"opportunity":272},"融合无人机多源遥感特征与深度学习，高精度估算茶树生物量与氮素积累量。","融合光谱、纹理、谐波特征的多域融合指标，结合无人机数据与深度学习模型。","多域融合指数显著优于传统光谱指标，实现茶园生物量与氮素高精度反演。","可探索多域融合特征在其他作物或不同生长阶段的适用性，或结合时序数据预测氮素动态。","agent","2026-09-02T00:05:07.021728Z",{"id":276,"title":277,"url":278,"summary":279,"summary_zh":280,"content":9,"source_name":281,"source_url":278,"published_at":282,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":283,"score_detail":284,"sources":286,"tags":288,"primary_topic":291,"secondary_topics":296,"classification":298,"search_phrases":316,"slug":319,"view_count":118,"doi":320,"paper":321,"created_at":337},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","2026-10-18T00:00:00Z",64,{"impact":17,"substance":141,"depth":139,"authority":13,"freshness":59,"relevant":22,"comment":285},"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。",[287],{"name":281,"url":278},[27,28,29,289,290],"番茄病害","图像识别",{"code":292,"slug":293,"name":294,"path":295,"parent_code":37},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",[297],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},{"status":46,"locked":14,"version":47,"method":48,"evidence":299},{"candidates":300,"agricultural_evidence":53},[301,307,311],{"code":292,"score":52,"supported":53,"title":302,"abstract":304,"tags":306},[303],"tomato",[305,303],"番茄",[289],{"code":40,"score":21,"supported":53,"title":308,"abstract":309,"tags":310},[],[29,62],[28,290,29],{"code":73,"score":166,"supported":53,"title":312,"abstract":313,"tags":315},[],[27,314],"smart agriculture",[27],[317,318],"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":320,"openalex_id":322,"authors":323,"venue":281,"cited_by_count":118,"oa_url":329,"card":330,"direction":336,"ingested_from":125},"W7222977034",[324,326],{"name":325,"orcid":9},"Mrs. Vijetalaxmi A Hegde",{"name":327,"orcid":328},"Abhay Ramakant Kasetwar","https:\u002F\u002Forcid.org\u002F0000-0002-5800-8764","https:\u002F\u002Factainnovations.com\u002Findex.php\u002Fpub\u002Farticle\u002Fdownload\u002F1918\u002F981",{"tldr":331,"method":332,"finding":333,"direction":334,"opportunity":335},"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。","农业人工智能与决策模型","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。","智慧农业 \u002F 农业物联网","2026-10-20T23:30:10.424511Z",{"id":339,"title":340,"url":341,"summary":342,"summary_zh":343,"content":9,"source_name":344,"source_url":341,"published_at":282,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":345,"score_detail":346,"sources":348,"tags":350,"primary_topic":353,"secondary_topics":354,"classification":355,"search_phrases":379,"slug":382,"view_count":118,"doi":383,"paper":384,"created_at":392},5385,"A Comprehensive Review of Deep Learning and Computer Vision Techniques for Fruit Quality Assessment and Classification","https:\u002F\u002Fdoi.org\u002F10.25258\u002Fijddt.16.75s.151","Fruit image analysis has become an essential research area in precision agriculture, enabling automated fruit classification, quality grading, maturity estimation, disease diagnosis, yield prediction, and intelligent harvesting through advances in computer vision and artificial intelligence (AI).Recent progress in imaging technologies, large-scale benchmark datasets, and deep learning has substantially improved the accuracy, robustness, and scalability of automated fruit analysis systems.This survey presents a comprehensive and systematic review of image processing and AI techniques for fruit image analysis.It covers the complete image analysis pipeline, including image acquisition, preprocessing, enhancement, color space transformation, segmentation, feature extraction, and handcrafted texture descriptors.Furthermore, conventional machine learning algorithms are reviewed alongside state-of-the-art deep learning architectures, including Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), and Vision Mamba models, for applications such as fruit recognition, quality assessment, disease detection, ripeness estimation, and object detection.Publicly available benchmark datasets, evaluation metrics, and performance comparison strategies are also summarized to facilitate objective assessment of existing approaches.In addition, the survey discusses key research challenges, including illumination variations, occlusion, complex orchard environments, limited annotated datasets, computational complexity, and domain adaptation, while highlighting emerging research trends such as self-supervised learning, multimodal learning, explainable AI, lightweight deep networks, foundation models, and edge intelligence for real-time deployment.By providing a unified review of methodologies, datasets, evaluation protocols, challenges, and future research directions, this survey offers a comprehensive reference for researchers and practitioners developing intelligent and scalable fruit image analysis systems for precision agriculture.","水果图像分析已成为精准农业中的一个重要研究领域，借助计算机视觉和人工智能（AI）的进步，实现了果实自动分类、品质分级、成熟度估计、病害诊断、产量预测和智能采收。成像技术、大规模基准数据集和深度学习的最新进展显著提升了自动化水果分析系统的准确性、鲁棒性和可扩展性。本综述对用于水果图像分析的图像处理和AI技术进行了全面而系统的回顾。内容涵盖完整的图像分析流程，包括图像采集、预处理、增强、颜色空间变换、分割、特征提取和手工纹理描述符。此外，综述了传统机器学习算法以及最先进的深度学习架构，包括卷积神经网络（CNN）、视觉Transformer（ViT）和视觉Mamba模型，用于果实识别、品质评估、病害检测、成熟度估计和目标检测等应用。同时总结了公开可用的基准数据集、评价指标和性能比较策略，以便对现有方法进行客观评估。此外，本综述讨论了关键研究挑战，包括光照变化、遮挡、复杂果园环境、标注数据集有限、计算复杂度和域适应，并重点介绍了新兴研究趋势，如自监督学习、多模态学习、可解释AI、轻量级深度网络、基础模型和面向实时部署的边缘智能。通过提供方法、数据集、评价协议、挑战和未来研究方向的统一综述，本综述为开发面向精准农业的智能且可扩展的水果图像分析系统的研究人员和实践者提供了全面的参考。","International Journal of Drug Delivery Technology",56,{"impact":59,"substance":141,"depth":139,"authority":66,"freshness":59,"relevant":22,"comment":347},"系统性综述覆盖CNN、ViT、Vision Mamba等前沿方法，信息量足，但期刊权威性一般且偏学术综述，公共价值有限，适合主题聚合而非每日精选头条。",[349],{"name":344,"url":341},[27,28,351,29,352],"计算机视觉","水果品质检测",{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},[],{"status":46,"locked":14,"version":47,"method":48,"evidence":356},{"candidates":357,"agricultural_evidence":53},[358,368,372],{"code":40,"score":359,"supported":53,"title":360,"abstract":362,"tags":367},28,[62,361],"computer vision",[363,29,364,351,365,62,366,361],"目标检测","机器学习","machine learning","artificial intelligence",[28,29,351],{"code":73,"score":166,"supported":53,"title":369,"abstract":370,"tags":371},[],[77,169],[27],{"code":373,"score":374,"supported":53,"title":375,"abstract":376,"tags":378},"B2",3,[],[377],"orchard",[],[380,381],"水果品质 计算机视觉 深度学习","农业人工智能 水果品质检测 计算机视觉 智慧农业","水果品质计算机视觉深度学习-5385","10.25258\u002Fijddt.16.75s.151",{"doi":383,"openalex_id":385,"authors":386,"venue":344,"cited_by_count":118,"oa_url":341,"card":387,"direction":336,"ingested_from":125},"W7222892831",[],{"tldr":388,"method":389,"finding":390,"direction":123,"opportunity":391},"系统综述深度学习与计算机视觉在水果品质评估与分级中的全流程方法。","综述图像采集到分割、特征提取及CNN、ViT、Vision Mamba等模型与公","深度学习显著提升水果分析精度与鲁棒性，但光照、遮挡和小样本仍是主要瓶颈。","可探索自监督与轻量化模型在复杂果园遮挡、小样本场景下的边缘实时部署。","2026-10-20T23:30:10.066042Z",{"id":394,"title":395,"url":396,"summary":397,"summary_zh":398,"content":9,"source_name":399,"source_url":396,"published_at":400,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":401,"score_detail":402,"sources":404,"tags":406,"primary_topic":409,"secondary_topics":415,"classification":417,"search_phrases":438,"slug":441,"view_count":22,"doi":442,"paper":443,"created_at":476},5164,"Lightweight Hybrid CNN-Transformer Network for Robust Maize Leaf Disease Detection and Edge Deployment","https:\u002F\u002Fdoi.org\u002F10.22266\u002Fijies2026.1130.50","Early and accurate detection of maize leaf diseases is essential for improving crop productivity and supporting precision agriculture.However, deep learning models trained on controlled leaf-image datasets often experience performance degradation under real-world agricultural conditions because of illumination variation, background clutter, occlusion, leaf orientation, and imaging noise.To address these challenges, this study develops a lightweight deep learning framework for robust maize leaf disease classification and edge-oriented deployment.The framework employs MobileNetV3-Small for efficient extraction of disease-discriminative visual features and investigates lightweight feature-learning strategies for improving classification under challenging field conditions.Experiments were conducted using PlantVillage-Maize and Field-Maize datasets comprising 6,188 images from four categories: Maize Leaf Blight (MLB), Common Rust, Healthy, and Leaf Spot.The image-only MobileNetV3-Small model achieved 97.8% accuracy and a 97.4% F1-score on the controlled PlantVillage-Maize dataset, while its performance decreased to 86.1% accuracy and an 83.9% F1-score on Field-Maize, highlighting the domain gap between controlled and real-field imagery.Robustness was further evaluated under Gaussian noise, illumination variation, and partial occlusion, together with computational-efficiency and edge-deployment analyses.The findings demonstrate the importance of lightweight and robust visual feature learning for practical maize disease diagnosis under real-world agricultural imaging conditions.","早期准确检测玉米叶片病害对于提高作物生产力和支持精准农业至关重要。然而，在受控叶片图像数据集上训练的深度学习模型，在真实农业条件下往往会出现性能下降，原因包括光照变化、背景杂乱、遮挡、叶片朝向以及成像噪声。为应对这些挑战，本研究开发了一个轻量级深度学习框架，用于稳健的玉米叶片病害分类和面向边缘的部署。该框架采用MobileNetV3-Small高效提取具有病害判别力的视觉特征，并研究轻量级特征学习策略，以提升在复杂田间条件下的分类性能。实验使用PlantVillage-Maize和Field-Maize数据集进行，共包含来自四个类别的6，188张图像：玉米大斑病（Maize Leaf Blight，MLB）、普通锈病（Common Rust）、健康（Healthy）和叶斑病（Leaf Spot）。仅使用图像的MobileNetV3-Small模型在受控的PlantVillage-Maize数据集上达到了97.8%的准确率和97.4%的F1分数，而在Field-Maize上的性能下降至86.1%的准确率和83.9%的F1分数，凸显了受控图像与真实田间图像之间的域差距。研究进一步在高斯噪声、光照变化和部分遮挡条件下评估了稳健性，并进行了计算效率和边缘部署分析。研究结果表明，轻量级且稳健的视觉特征学习对于真实农业成像条件下的实际玉米病害诊断具有重要意义。","International journal of intelligent engineering and systems","2026-10-16T00:00:00Z",69,{"impact":17,"substance":18,"depth":139,"authority":17,"freshness":21,"relevant":22,"comment":403},"该研究提出轻量级玉米叶病检测框架，并揭示受控数据集与田间数据之间的性能差距，方法扎实、结论有参考价值，但属细分技术论文，公共影响有限。",[405],{"name":399,"url":396},[27,28,407,29,408],"边缘计算","玉米病害",{"code":410,"slug":411,"name":412,"path":413,"parent_code":414},"A3","maize","玉米","\u002Ftopic\u002Ffield-crops\u002Fmaize","A",[416],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},{"status":46,"locked":14,"version":47,"method":48,"evidence":418},{"candidates":419,"agricultural_evidence":53},[420,425,429,433],{"code":410,"score":421,"supported":53,"title":422,"abstract":423,"tags":424},14,[411],[412,411],[],{"code":40,"score":59,"supported":53,"title":426,"abstract":427,"tags":428},[],[29,62],[28,29],{"code":73,"score":166,"supported":53,"title":430,"abstract":431,"tags":432},[],[77,169],[27],{"code":434,"score":22,"supported":14,"title":435,"abstract":436,"tags":437},"I5",[],[],[407],[439,440],"MobileNetV3 玉米叶部病害","玉米叶片病害 边缘部署","MobileNetV3玉米叶部病害-5164","10.22266\u002Fijies2026.1130.50",{"doi":442,"openalex_id":444,"authors":445,"venue":399,"cited_by_count":118,"oa_url":396,"card":471,"direction":123,"ingested_from":125},"W7222602850",[446,449,452,454,456,459,461,463,466,468],{"name":447,"orcid":448},"Awwab Mohammad","https:\u002F\u002Forcid.org\u002F0000-0002-1148-5678",{"name":450,"orcid":451},"Punith kumar","https:\u002F\u002Forcid.org\u002F0009-0005-5676-4194",{"name":453,"orcid":9},"Bhavani K",{"name":455,"orcid":9},"Sailee N. Panchbhai",{"name":457,"orcid":458},"Sagarkumar S. Badhiye","https:\u002F\u002Forcid.org\u002F0000-0002-0710-3761",{"name":460,"orcid":9},"Manoj Kumar S B",{"name":462,"orcid":9},"Dharamvir",{"name":464,"orcid":465},"Sumit Kumar Sar","https:\u002F\u002Forcid.org\u002F0000-0003-0384-9088",{"name":467,"orcid":9},"Rani Chhagan Dalvi",{"name":469,"orcid":470},"Yogesh Harishchandra Bhosale","https:\u002F\u002Forcid.org\u002F0000-0001-6901-1419",{"tldr":472,"method":473,"finding":474,"direction":334,"opportunity":475},"构建轻量MobileNetV3-Small框架实现玉米叶片病害分类并评估边缘部署。","MobileNetV3-Small，PlantVillage-Maize与Fie","受控数据准确率97.8%，田间降至86.1%，揭示明显域差距。","可研究跨域自适应与田间鲁棒特征学习，缩小受控与真实场景性能差距。","2026-10-17T23:30:28.248539Z",{"id":478,"title":479,"url":480,"summary":481,"summary_zh":482,"content":9,"source_name":10,"source_url":480,"published_at":400,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":483,"score_detail":484,"sources":487,"tags":489,"primary_topic":492,"secondary_topics":497,"classification":499,"search_phrases":515,"slug":518,"view_count":22,"doi":519,"paper":520,"created_at":554},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。",76,{"impact":139,"substance":485,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":486},21,"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。",[488],{"name":10,"url":480},[27,28,29,490,491],"病虫害识别","烟草",{"code":493,"slug":494,"name":495,"path":496,"parent_code":414},"A6","cash-crops","棉麻糖与特色经作","\u002Ftopic\u002Ffield-crops\u002Fcash-crops",[498],{"code":40,"slug":41,"name":42,"path":43,"parent_code":44},{"status":46,"locked":14,"version":47,"method":48,"evidence":500},{"candidates":501,"agricultural_evidence":53},[502,507,511],{"code":493,"score":52,"supported":53,"title":503,"abstract":505,"tags":506},[504],"tobacco",[491,504],[491],{"code":40,"score":21,"supported":53,"title":508,"abstract":509,"tags":510},[],[29,62],[28,29,490],{"code":73,"score":166,"supported":53,"title":512,"abstract":513,"tags":514},[],[27,314],[27],[516,517],"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":519,"openalex_id":521,"authors":522,"venue":10,"cited_by_count":118,"oa_url":548,"card":549,"direction":336,"ingested_from":125},"W7222603139",[523,525,527,529,531,533,535,537,540,543,546],{"name":524,"orcid":9},"Shun He",{"name":526,"orcid":9},"Sifan Li",{"name":528,"orcid":9},"Xinjian Li",{"name":530,"orcid":9},"Qianqian Luo",{"name":532,"orcid":9},"Dong Sufen",{"name":534,"orcid":9},"Lu Lei",{"name":536,"orcid":9},"Xingjiang Chen",{"name":538,"orcid":539},"Liuti Cai","https:\u002F\u002Forcid.org\u002F0000-0002-9151-8002",{"name":541,"orcid":542},"Jian‐Yu Meng","https:\u002F\u002Forcid.org\u002F0000-0002-5796-1086",{"name":544,"orcid":545},"Ning Lu","https:\u002F\u002Forcid.org\u002F0000-0003-0214-5038",{"name":547,"orcid":9},"Hancheng Wang","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fplant-science\u002Farticles\u002F10.3389\u002Ffpls.2026.1871587\u002Fpdf",{"tldr":550,"method":551,"finding":552,"direction":334,"opportunity":553},"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"]