[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3067":3,"related-3067":97},{"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":39,"search_phrases":66,"slug":69,"view_count":70,"doi":71,"paper":72,"created_at":96},3067,"Automated detection of visible venation patterns in cowpea leaves under agricultural engineering","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.sasc.2026.200636","Accurate assessment of crop leaf characteristics is important for automated crop phenotyping and precision agriculture. However, manual assessment of leaf characteristics is time-consuming and may be affected by subjective interpretation. This study proposes an RGB-based convolutional neural network (CNN) framework for automated classification of surface-visible venation patterns in cowpea (Vigna unguiculata) leaves. A dataset of 2500 RGB cowpea leaf images, comprising Healthy and Structurally Stressed samples, was used to develop and evaluate the proposed framework. The images were preprocessed using standard image-processing operations and subsequently analyzed using a CNN for binary classification of the two visible-venation categories. Canny edge detection was additionally employed as a supplementary visualization technique to highlight surface-visible venation patterns and support qualitative interpretation; it was not used as the primary classification mechanism. The proposed CNN achieved 95.4% accuracy, 94.8% precision, 95.1% recall, and 95.0% F1-score on the evaluated test dataset. Comparative experiments with conventional feature-based approaches, including GLCM, SIFT, and HOG, showed lower observed classification performance. The results demonstrate the potential of RGB-based deep learning for automated analysis of surface-visible cowpea leaf venation patterns and provide a non-destructive image-based approach for crop phenotyping. The proposed framework can support scalable agricultural image analysis while avoiding claims of direct Surface-visible vein pattern anatomical characterization. The dataset is available at this link: “ https:\u002F\u002Fgithub.com\u002Fvijayachoudharyresearcher\u002FComputer-Vision–Cowpea-Leaves-Internal-Structure “.","准确评估作物叶片特征对于自动化作物表型分析和精准农业具有重要意义。然而，人工评估叶片特征耗时较长，且可能受到主观解释的影响。本研究提出了一种基于RGB的卷积神经网络（CNN）框架，用于对豇豆（Vigna unguiculata）叶片表面可见脉纹模式进行自动分类。研究使用包含健康和结构胁迫样本的2500张RGB豇豆叶片图像数据集来开发和评估所提出的框架。图像经标准图像处理操作预处理后，利用CNN对两个可见脉纹类别进行二分类分析。此外，研究还采用Canny边缘检测作为补充可视化技术，以突出表面可见脉纹模式并支持定性解释；其并未被用作主要分类机制。所提出的CNN在评估测试数据集上取得了95.4%的准确率、94.8%的精确率、95.1%的召回率和95.0%的F1分数。与传统基于特征的方法（包括GLCM、SIFT和HOG）的比较实验显示，其分类性能较低。结果表明，基于RGB的深度学习在自动化分析豇豆叶片表面可见脉纹模式方面具有潜力，并为作物表型分析提供了一种基于图像的非破坏性方法。所提出的框架可支持可扩展的农业图像分析，同时避免声称对表面可见脉纹模式进行直接的解剖学表征。数据集可通过以下链接获取：“https:\u002F\u002Fgithub.com\u002Fvijayachoudharyresearcher\u002FComputer-Vision–Cowpea-Leaves-Internal-Structure”。",null,"Systems and Soft Computing","2026-09-18T00:00:00Z","论文",10,false,69,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},12,20,16,13,8,1,"基于RGB与CNN的豇豆叶脉自动识别研究，方法清晰、数据规模可观，对作物表型与精准农业有参考价值，但属细分领域技术进展，影响范围有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","作物表型","图像识别","豇豆",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":40,"locked":14,"version":41,"method":42,"evidence":43},"classified","2026-10-22.2","rules",{"candidates":44,"agricultural_evidence":46},[45,52,60],{"code":33,"score":21,"supported":46,"title":47,"abstract":48,"tags":51},true,[],[49,50],"深度学习","deep learning",[28,30],{"code":53,"score":54,"supported":46,"title":55,"abstract":56,"tags":59},"I1",7,[],[57,58],"精准农业","precision agriculture",[27],{"code":61,"score":54,"supported":46,"title":62,"abstract":63,"tags":65},"I3",[],[29,64],"crop phenotyping",[29],[67,68],"豇豆叶片 叶脉 识别","RGB CNN 作物表型","豇豆叶片叶脉识别-3067",2,"10.1016\u002Fj.sasc.2026.200636",{"doi":71,"openalex_id":73,"authors":74,"venue":10,"cited_by_count":88,"oa_url":6,"card":89,"direction":93,"ingested_from":95},"W7213555214",[75,78,81,83,85],{"name":76,"orcid":77},"Vijaya Choudhary","https:\u002F\u002Forcid.org\u002F0000-0003-3417-1943",{"name":79,"orcid":80},"Paramita Guha","https:\u002F\u002Forcid.org\u002F0000-0001-6696-4812",{"name":82,"orcid":9},"Jai Prakash Mishra",{"name":84,"orcid":9},"Himanshu Sharma",{"name":86,"orcid":87},"Munish Sabharwal","https:\u002F\u002Forcid.org\u002F0000-0002-7338-6982",0,{"tldr":90,"method":91,"finding":92,"direction":93,"opportunity":94},"用RGB卷积神经网络自动分类豇豆叶片可见脉纹，区分健康与结构胁迫样本。","2500张RGB豇豆叶图像，CNN二分类，Canny边缘检测辅助可视化，对比GL","CNN达95.4%准确率，优于传统特征方法，可实现无损作物表型分析。","农业遥感与作物表型","可扩展至多作物、多胁迫类型及田间实时检测，并结合多光谱提升脉纹表型解析。","openalex","2026-09-21T23:30:18.575016Z",{"total":98,"page":22,"page_size":98,"items":99},6,[100,166,227,281,346,405],{"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":112,"tags":114,"primary_topic":116,"secondary_topics":122,"classification":124,"search_phrases":144,"slug":147,"view_count":88,"doi":148,"paper":149,"created_at":165},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":110,"depth":19,"authority":13,"freshness":21,"relevant":22,"comment":111},18,"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。",[113],{"name":106,"url":103},[27,28,49,115,30],"番茄病害",{"code":117,"slug":118,"name":119,"path":120,"parent_code":121},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables","B",[123],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":125},{"candidates":126,"agricultural_evidence":46},[127,134,139],{"code":117,"score":128,"supported":46,"title":129,"abstract":131,"tags":133},15,[130],"tomato",[132,130],"番茄",[115],{"code":33,"score":135,"supported":46,"title":136,"abstract":137,"tags":138},9,[],[49,50],[28,30,49],{"code":53,"score":54,"supported":46,"title":140,"abstract":141,"tags":143},[],[27,142],"smart agriculture",[27],[145,146],"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":148,"openalex_id":150,"authors":151,"venue":106,"cited_by_count":88,"oa_url":157,"card":158,"direction":164,"ingested_from":95},"W7222977034",[152,154],{"name":153,"orcid":9},"Mrs. Vijetalaxmi A Hegde",{"name":155,"orcid":156},"Abhay Ramakant Kasetwar","https:\u002F\u002Forcid.org\u002F0000-0002-5800-8764","https:\u002F\u002Factainnovations.com\u002Findex.php\u002Fpub\u002Farticle\u002Fdownload\u002F1918\u002F981",{"tldr":159,"method":160,"finding":161,"direction":162,"opportunity":163},"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":167,"title":168,"url":169,"summary":170,"summary_zh":171,"content":9,"source_name":172,"source_url":169,"published_at":173,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":174,"score_detail":175,"sources":177,"tags":179,"primary_topic":182,"secondary_topics":183,"classification":184,"search_phrases":209,"slug":212,"view_count":22,"doi":213,"paper":214,"created_at":226},4833,"From Handcrafted Descriptors to Deep Embeddings: A Comparative Study for Regional Hazelnut Image Classification","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1946462","Automated regional hazelnut image classification is relevant for agricultural traceability, origin aware sorting, and decision support in food processing systems. This study presents a unified comparative benchmark for three class hazelnut image classification using Van, Ordu, and Giresun categories. The experimental framework integrates classical handcrafted descriptors, deep transfer learning models, and reduced space hybrid classical and quantum compatible classifiers. In the classical branch, HOG and RGB color histogram features were combined and evaluated using KNN and SVM based classifiers. In the deep learning branch, ResNet50 and EfficientNetB0 were trained through a two stage transfer learning strategy, and Grad-CAM was employed to provide qualitative interpretability. In the hybrid branch, 1280 dimensional embeddings extracted from the GAP layer of the best EfficientNetB0 model were compressed into low dimensional spaces for reduced space SVC and QSVC experiments. Under the present image level split protocol, the strongest handcrafted baseline, PCA + SVM, achieved 98.90% accuracy and 0.9888 macro-F1, while ResNet50 reached 99.45% accuracy and 0.9945 macro-F1. EfficientNetB0 achieved 100% test accuracy, while the reduced space classical SVC trained on compressed EfficientNetB0 embeddings attained 99.63% accuracy and 0.9961 macro F1 on the full test set. Under the fair subset protocol, classical SVC achieved 98.33% accuracy, whereas the best QSVC configuration reached 90.83% accuracy. This result should be interpreted as a strong within dataset benchmark rather than definitive evidence of object level or deployment level generalization. Grad-CAM visualizations further suggested that the deep models focused primarily on semantically meaningful hazelnut regions rather than on background patterns. The findings indicate that representation quality is a central factor in this task, while the reported near ceiling performance should be interpreted in light of the image level partitioning strategy","区域榛子图像的自动化分类对于农业可追溯性、产地感知分选以及食品加工系统中的决策支持具有重要意义。本研究提出了一个统一的比较基准，用于对Van、Ordu和Giresun三个类别的榛子图像进行分类。实验框架整合了经典手工设计描述子、深度迁移学习模型以及降维空间下的经典与量子兼容混合分类器。在经典分支中，将HOG与RGB颜色直方图特征进行组合，并使用基于KNN和SVM的分类器进行评估。在深度学习分支中，通过两阶段迁移学习策略训练ResNet50和EfficientNetB0，并采用Grad-CAM提供定性可解释性。在混合分支中，将从最佳EfficientNetB0模型GAP层提取的1280维嵌入压缩至低维空间，用于降维空间SVC和QSVC实验。在当前图像级划分协议下，最强的手工特征基线PCA + SVM达到了98.90%的准确率和0.9888的宏F1值，而ResNet50达到了99.45%的准确率和0.9945的宏F1值。EfficientNetB0达到了100%的测试准确率，而在压缩后的EfficientNetB0嵌入上训练的降维空间经典SVC在完整测试集上达到了99.63%的准确率和0.9961的宏F1值。在公平子集协议下，经典SVC达到了98.33%的准确率，而最佳QSVC配置达到了90.83%的准确率。该结果应被解释为数据集内的强基准，而非对象级或部署级泛化能力的确定性证据。Grad-CAM可视化进一步表明，深度模型主要关注具有语义意义的榛子区域，而非背景模式。研究结果表明，表征质量是该任务的核心因素，而所报告的接近上限的性能应结合图像级划分策略来理解。","Tarım Bilimleri Dergisi","2026-10-12T00:00:00Z",63,{"impact":21,"substance":110,"depth":19,"authority":17,"freshness":135,"relevant":22,"comment":176},"方法对比扎实、结论可靠，但属数据集内基准研究，产业落地价值有限，可作技术参考而非每日精选头条。",[178],{"name":172,"url":169},[27,28,180,30,181],"农产品溯源","榛子",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":185},{"candidates":186,"agricultural_evidence":46},[187,193,200,205],{"code":33,"score":188,"supported":46,"title":189,"abstract":190,"tags":192},11,[],[49,50,191],"decision support",[28,30],{"code":194,"score":98,"supported":46,"title":195,"abstract":196,"tags":199},"E1",[],[197,198],"食品加工","food processing",[],{"code":201,"score":22,"supported":14,"title":202,"abstract":203,"tags":204},"E4",[],[],[180],{"code":53,"score":22,"supported":14,"title":206,"abstract":207,"tags":208},[],[],[27],[210,211],"榛子 产地 图像分类","EfficientNetB0 榛子 识别","榛子产地图像分类-4833","10.15832\u002Fankutbd.1946462",{"doi":213,"openalex_id":215,"authors":216,"venue":172,"cited_by_count":88,"oa_url":220,"card":221,"direction":162,"ingested_from":95},"W7221535543",[217],{"name":218,"orcid":219},"Özgür Tomak","https:\u002F\u002Forcid.org\u002F0000-0003-2993-6913","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5990175",{"tldr":222,"method":223,"finding":224,"direction":162,"opportunity":225},"比较手工特征、深度迁移学习和量子兼容分类器对三个地区榛子图像的分类性能。","HOG+颜色直方图、ResNet50\u002FEfficientNetB0迁移学习、Gr","EfficientNetB0达100%准确率，深度嵌入质量是关键，但结果受图像级划分限制。","可探索对象级划分与跨域泛化，并验证量子分类器在真实农业溯源场景的实用性。","2026-10-13T23:30:38.487402Z",{"id":228,"title":229,"url":230,"summary":231,"summary_zh":232,"content":9,"source_name":233,"source_url":230,"published_at":234,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":174,"score_detail":235,"sources":237,"tags":239,"primary_topic":241,"secondary_topics":242,"classification":243,"search_phrases":254,"slug":257,"view_count":22,"doi":258,"paper":259,"created_at":280},4743,"Image Processing-Based Efficient Hybrid Feature Extraction Approach for Plant Leaf Disease Recognition","https:\u002F\u002Fdoi.org\u002F10.14445\u002F23488549\u002Fijece-v13i9p107","In this paper, a novel hybrid method for plant leaf disease recognition is presented, combining image processing techniques with Deep Learning and handcrafted feature analysis. Leaf images are processed using a Convolutional Neural Network (CNN) to automatically extract hierarchical features, while handcrafted features based on leaf shape, color, and texture provide complementary disease-related information. The combined feature set is fed into a multi-class classifier to accurately identify specific disease types. Evaluation on a diverse dataset of healthy and diseased leaves demonstrates an average accuracy of 98%, outperforming approaches that rely solely on deep learning or handcrafted features. The results highlight the effectiveness of integrating deep and handcrafted features, offering a reliable, scalable solution for precision agriculture and serving as a foundation for future research in automated plant disease detection.","本文提出了一种新颖的植物叶片病害识别混合方法，将图像处理技术与深度学习及手工特征分析相结合。利用卷积神经网络（CNN）对叶片图像进行处理，以自动提取层次化特征，同时基于叶片形状、颜色和纹理的手工特征提供了与病害相关的补充信息。将组合后的特征集输入多类分类器，以准确识别具体的病害类型。在包含健康叶片和患病叶片的多样化数据集上进行评估，结果表明平均准确率达到98%，优于仅依赖深度学习或手工特征的方法。结果凸显了融合深度特征与手工特征的有效性，为精准农业提供了一种可靠且可扩展的解决方案，并为自动化植物病害检测的未来研究奠定了基础。","International Journal of Electronics and Communication Engineering","2026-10-10T00:00:00Z",{"impact":17,"substance":110,"depth":128,"authority":13,"freshness":21,"relevant":22,"comment":236},"提出深度学习与手工特征融合的叶片病害识别方法，准确率达98%，对精准农业有参考价值，但属实验室论文，产业影响有限。",[238],{"name":233,"url":230},[27,28,240,30],"植物病害识别",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":244},{"candidates":245,"agricultural_evidence":46},[246,250],{"code":33,"score":135,"supported":46,"title":247,"abstract":248,"tags":249},[],[49,50],[28,30,240],{"code":53,"score":54,"supported":46,"title":251,"abstract":252,"tags":253},[],[57,58],[27],[255,256],"植物叶片病害识别 图像处理","CNN 特征提取 病害检测","植物叶片病害识别图像处理-4743","10.14445\u002F23488549\u002Fijece-v13i9p107",{"doi":258,"openalex_id":260,"authors":261,"venue":233,"cited_by_count":88,"oa_url":274,"card":275,"direction":164,"ingested_from":95},"W7221322423",[262,264,267,270,272],{"name":263,"orcid":9},"Arundhati Bora",{"name":265,"orcid":266},"Parismita Sarma","https:\u002F\u002Forcid.org\u002F0000-0003-1904-820X",{"name":268,"orcid":269},"Dankan Gowda V","https:\u002F\u002Forcid.org\u002F0000-0003-0724-0333",{"name":271,"orcid":9},"Manash Pratim Bhuyan",{"name":273,"orcid":9},"Jumi Sarmah","https:\u002F\u002Fwww.internationaljournalssrg.org\u002F..\u002FIJECE\u002F2026\u002FVolume13-Issue9\u002FIJECE-V13I9P107.pdf",{"tldr":276,"method":277,"finding":278,"direction":162,"opportunity":279},"提出融合CNN深度特征与手工特征的方法，用于植物叶片病害识别。","CNN自动提取特征，结合叶片形状、颜色、纹理手工特征，多类分类器。","平均准确率达98%，优于仅用深度学习或手工特征的方法。","可探索轻量化混合特征模型在移动端或边缘设备的实时病害识别应用。","2026-10-12T23:30:07.942335Z",{"id":282,"title":283,"url":284,"summary":285,"summary_zh":286,"content":9,"source_name":287,"source_url":284,"published_at":288,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":289,"score_detail":290,"sources":293,"tags":295,"primary_topic":297,"secondary_topics":298,"classification":300,"search_phrases":317,"slug":320,"view_count":22,"doi":321,"paper":322,"created_at":345},4494,"Tomato ripeness classification with weight-based and texture-adaptive pooling in CNNs","https:\u002F\u002Fdoi.org\u002F10.64960\u002Feasr.2026.266223","This research introduces a weight-based mixed pooling and adaptive fuzzy pooling mechanism based on feature map texture analysis for attention mechanism Squeeze-and-Excitation Networks (SE) and Convolutional Neural Networks (CNNs) in tomato ripeness classification. Traditional CNNs and SE networks rely on deterministic pooling strategies that can be suboptimal for capturing diverse visual features such as texture and color gradients in fruit. A novel fuzzy pooling system is proposed that dynamically adjusts the fusion ratio of min and max pooling based on real-time texture analysis of feature maps via a Laplacian filter. This mechanism was integrated into Darknet53, ResNet34, and ResNet101 architectures, both directly and within the SE attention block. The method was evaluated on a five-class tomato ripeness dataset using 10-fold cross-validation against traditional and fixed-weight pooling methods. The fuzzy and mixed pooling approach demonstrated superior performance on certain models which include ESEnet: Darknet53 + SE (Fuzzy pool) achieved an average F1 score across folds of 0.8429, outperforming Darknet53 + baseline SE (Average pool) at 0.8350; Extended Net: ResNet34 + Fuzzy pool achieved an average F1 score across folds of 0.8305, outperforming baseline ResNet34 at 0.8130 and ResNet101 + Fuzzy pool achieved an average F1 score across folds of 0.8284, outperforming baseline ResNet101 at 0.8267. This improvement was realized with minimal increase in computational overhead. Specifically, measurements of model parameters (M) and FLOPs (GMACs) showed only a slight increase across all architectures after integrating SEnet into the system, and zero increase after integrating the Laplacian filter and fuzzy logic. The study concludes that logic-based adaptive pooling is a robust strategy that improves feature representation in CNNs, achieving a balance between classification accuracy and computational cost. These findings offer practical value for developing more reliable automated harvesting systems and provide a generalizable enhancement for a wide range of computer vision classification tasks.","本研究提出了一种基于特征图纹理分析的权重混合池化与自适应模糊池化机制，用于番茄成熟度分类中的注意力机制挤压激励网络（Squeeze-and-Excitation Networks, SE）和卷积神经网络（Convolutional Neural Networks, CNNs）。传统的CNN和SE网络依赖确定性池化策略，在捕捉果实纹理和颜色梯度等多样化视觉特征时可能并非最优。本文提出了一种新颖的模糊池化系统，通过拉普拉斯滤波器对特征图进行实时纹理分析，动态调整最小池化和最大池化的融合比例。该机制被集成到Darknet53、ResNet34和ResNet101架构中，既可直接集成，也可嵌入SE注意力模块内。该方法在五类番茄成熟度数据集上采用10折交叉验证，与传统池化及固定权重池化方法进行了对比评估。模糊池化和混合池化方法在部分模型上表现出优越性能，包括：ESEnet：Darknet53 + SE（模糊池化）在各折上的平均F1得分为0.8429，优于Darknet53 + 基线SE（平均池化）的0.8350；扩展网络：ResNet34 + 模糊池化在各折上的平均F1得分为0.8305，优于基线ResNet34的0.8130；ResNet101 + 模糊池化在各折上的平均F1得分为0.8284，优于基线ResNet101的0.8267。这一改进仅带来了极小的计算开销增加。具体而言，模型参数量（M）和浮点运算次数（GMACs）的测量结果表明，将SEnet集成到系统后所有架构仅有轻微增加，而集成拉普拉斯滤波器和模糊逻辑后则无任何增加。研究结论认为，基于逻辑的自适应池化是一种鲁棒策略，能够改善CNN中的特征表示，在分类精度和计算成本之间实现平衡。这些发现为开发更可靠的自动化收获系统提供了实用价值，并为广泛的计算机视觉分类任务提供了一种可推广的增强方法。","Engineering and Applied Science Research","2026-10-07T00:00:00Z",66,{"impact":21,"substance":18,"depth":291,"authority":17,"freshness":135,"relevant":22,"comment":292},17,"提出基于纹理自适应模糊池化的CNN改进方法，在番茄成熟度分类上取得小幅但稳定的F1提升且算力开销几乎不增，对自动采摘与农业视觉识别有参考价值，但属细分方法类论文，影响面有限。",[294],{"name":287,"url":284},[27,28,132,30,296],"自动采摘",{"code":117,"slug":118,"name":119,"path":120,"parent_code":121},[299],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":301},{"candidates":302,"agricultural_evidence":46},[303,307,313],{"code":117,"score":128,"supported":46,"title":304,"abstract":305,"tags":306},[130],[132,130],[132],{"code":33,"score":21,"supported":46,"title":308,"abstract":309,"tags":312},[],[310,311],"计算机视觉","computer vision",[28,30],{"code":53,"score":22,"supported":14,"title":314,"abstract":315,"tags":316},[],[],[27],[318,319],"番茄 成熟度 分类 卷积神经网络","Darknet53 ResNet 番茄 成熟度","番茄成熟度分类卷积神经网络-4494","10.64960\u002Feasr.2026.266223",{"doi":321,"openalex_id":323,"authors":324,"venue":287,"cited_by_count":88,"oa_url":339,"card":340,"direction":164,"ingested_from":95},"W7220840158",[325,327,329,331,334,336],{"name":326,"orcid":9},"Rut Yatigul",{"name":328,"orcid":9},"Wisanu Jitviriya",{"name":330,"orcid":9},"Eiji Hayashi",{"name":332,"orcid":333},"Aran Blattler","https:\u002F\u002Forcid.org\u002F0000-0002-8684-6239",{"name":335,"orcid":9},"Watcharin Tangsuksant",{"name":337,"orcid":338},"Teppakorn Sittiwanchai","https:\u002F\u002Forcid.org\u002F0000-0001-9653-8750","https:\u002F\u002Fph01.tci-thaijo.org\u002Findex.php\u002Feasr\u002Farticle\u002Fdownload\u002F266223\u002F176569",{"tldr":341,"method":342,"finding":343,"direction":162,"opportunity":344},"提出基于纹理自适应模糊池化机制，提升CNN番茄成熟度分类性能。","用Laplacian滤波分析特征图纹理，动态调整最大最小池化融合比，集成到Dar","模糊池化在多个模型上F1提升，且计算开销几乎不增加。","可探索自适应池化在其它果蔬成熟度或田间复杂光照下的泛化能力。","2026-10-09T23:30:09.993787Z",{"id":347,"title":348,"url":349,"summary":350,"summary_zh":351,"content":9,"source_name":352,"source_url":349,"published_at":353,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":354,"sources":357,"tags":359,"primary_topic":361,"secondary_topics":362,"classification":363,"search_phrases":379,"slug":382,"view_count":383,"doi":384,"paper":385,"created_at":404},3963,"MSO-Det: A multi-level synergistic optimization framework for robust spike object detection in field scenes","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112473","MSO-Det: A multi-level synergistic optimization framework for robust spike object detection in field scenes。Computers and Electronics in Agriculture","MSO-Det：一种用于田间场景中鲁棒脉冲目标检测的多级协同优化框架。《计算机与电子农业》","Computers and Electronics in Agriculture","2026-10-01T00:00:00Z",{"impact":17,"substance":110,"depth":19,"authority":355,"freshness":135,"relevant":22,"comment":356},14,"发表于农业信息领域核心期刊的穗部目标检测新方法，方法有创新但属细分技术进展，公共影响有限。",[358],{"name":352,"url":349},[27,28,360,29],"目标检测",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":364},{"candidates":365,"agricultural_evidence":46},[366,371,375],{"code":33,"score":367,"supported":46,"title":368,"abstract":369,"tags":370},5,[],[360],[28,360],{"code":53,"score":22,"supported":14,"title":372,"abstract":373,"tags":374},[],[],[27],{"code":61,"score":22,"supported":14,"title":376,"abstract":377,"tags":378},[],[],[29],[380,381],"MSO-Det 穗部检测","田间场景 穗目标检测","MSO-Det穗部检测-3963",3,"10.1016\u002Fj.compag.2026.112473",{"doi":384,"openalex_id":386,"authors":387,"venue":352,"cited_by_count":88,"oa_url":9,"card":399,"direction":162,"ingested_from":95},"W7215045657",[388,390,392,394,396],{"name":389,"orcid":9},"Yiqun Wang",{"name":391,"orcid":9},"Shuo Zhang",{"name":393,"orcid":9},"Xiaolong Liu",{"name":395,"orcid":9},"Ze Ji",{"name":397,"orcid":398},"Weikuan Jia","https:\u002F\u002Forcid.org\u002F0000-0001-6242-3269",{"tldr":400,"method":401,"finding":402,"direction":162,"opportunity":403},"提出MSO-Det多级协同优化框架，用于田间场景下鲁棒的穗部目标检测。","多级协同优化框架，结合田间场景图像进行穗目标检测。","该框架提升了田间穗检测的鲁棒性，但摘要未给出具体指标。","可探索多级优化在复杂田间光照与遮挡下的泛化能力及轻量化部署。","2026-10-02T23:30:01.492947Z",{"id":406,"title":407,"url":408,"summary":409,"summary_zh":410,"content":9,"source_name":411,"source_url":408,"published_at":412,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":413,"score_detail":414,"sources":416,"tags":418,"primary_topic":420,"secondary_topics":425,"classification":427,"search_phrases":455,"slug":458,"view_count":367,"doi":459,"paper":460,"created_at":493},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。","Frontiers in Plant Science","2026-09-21T00:00:00Z",71,{"impact":17,"substance":18,"depth":291,"authority":20,"freshness":135,"relevant":22,"comment":415},"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。",[417],{"name":411,"url":408},[27,28,49,419,29],"可可",{"code":421,"slug":422,"name":423,"path":424,"parent_code":121},"B3","tea-coffee","茶叶与咖啡","\u002Ftopic\u002Fhorticulture\u002Ftea-coffee",[426],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":428},{"candidates":429,"agricultural_evidence":46},[430,435,439,446,451],{"code":421,"score":128,"supported":46,"title":431,"abstract":433,"tags":434},[432],"cacao",[419,432],[419],{"code":33,"score":21,"supported":46,"title":436,"abstract":437,"tags":438},[],[49,50],[28,49],{"code":440,"score":98,"supported":46,"title":441,"abstract":442,"tags":445},"B7",[],[443,444],"温室","greenhouse",[],{"code":53,"score":447,"supported":46,"title":448,"abstract":449,"tags":450},4,[],[57],[27],{"code":61,"score":22,"supported":14,"title":452,"abstract":453,"tags":454},[],[],[29],[456,457],"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","10.3389\u002Ffpls.2026.1885906",{"doi":459,"openalex_id":461,"authors":462,"venue":411,"cited_by_count":88,"oa_url":408,"card":488,"direction":93,"ingested_from":95},"W7213901640",[463,465,468,470,472,474,477,480,483,485],{"name":464,"orcid":9},"Ezekiel Ahn",{"name":466,"orcid":467},"Eun-Sung Park","https:\u002F\u002Forcid.org\u002F0000-0001-6826-2865",{"name":469,"orcid":9},"Moon S. Kim",{"name":471,"orcid":9},"Hangi Kim",{"name":473,"orcid":9},"Lalit M. Kandpal",{"name":475,"orcid":476},"Sunchung Park","https:\u002F\u002Forcid.org\u002F0000-0002-7398-9476",{"name":478,"orcid":479},"Seunghyun Lim","https:\u002F\u002Forcid.org\u002F0000-0003-3023-4863",{"name":481,"orcid":482},"Lyndel W. Meinhardt","https:\u002F\u002Forcid.org\u002F0000-0001-8299-2629",{"name":484,"orcid":9},"Byoung-Kwan Cho",{"name":486,"orcid":487},"Insuck Baek","https:\u002F\u002Forcid.org\u002F0000-0003-1044-349X",{"tldr":489,"method":490,"finding":491,"direction":93,"opportunity":492},"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。","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。","2026-09-22T23:30:19.972949Z"]