[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4009":3,"related-4009":85},{"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":63,"slug":66,"view_count":67,"doi":68,"paper":69,"created_at":84},4009,"Plant disease classification using lightweight deep learning architectures","https:\u002F\u002Fdoi.org\u002F10.1177\u002F18758967261494075","Detecting early-stage pathogens of plant diseases is a challenging task due to the low pathogen density and the subtlety or absence of visible symptoms. If not identified at an early stage, plant diseases can progress rapidly, making them significantly harder to control. Therefore, real-time and in-situ monitoring of plants is essential for tracking diseases and enabling early intervention. Within the scope of smart agriculture applications, crop health can be assessed and productivity can be maintained through mobile devices. In this study, the performance of lightweight neural network architectures on various plant disease datasets was compared, targeting devices with limited memory resources such as mobile devices and embedded systems, which we consider potential candidates for edge deployment based on the measured accuracy-efficiency trade-offs. The architectures evaluated include SqueezeNet, MobileViT, EdgeNeXt, GhostNet, EfficientNet, and LeViT. The plant datasets used in the experiments were Plant Disease, Cassava, and Turmeric. Each model was evaluated using Accuracy, F1-Score, Precision, and Recall metrics. Among the three datasets, the best performance was achieved on the Turmeric dataset using the GhostNet architecture, with scores of 0.9978 (Accuracy), 0.9976 (F1-Score), 0.9976 (Precision), and 0.9976 (Recall), respectively.","检测植物病害的早期病原体是一项具有挑战性的任务，因为病原体密度低，且可见症状细微或缺失。若未能在早期识别，植物病害会迅速进展，从而使其控制难度显著增加。因此，对植物进行实时原位监测对于追踪病害并实现早期干预至关重要。在智慧农业应用范围内，可通过移动设备评估作物健康并维持生产力。本研究比较了轻量级神经网络架构在不同植物病害数据集上的性能，目标针对内存资源受限的设备，如移动设备和嵌入式系统；基于所测得的精度-效率权衡，我们认为这些设备是边缘部署的潜在候选。所评估的架构包括SqueezeNet、MobileViT、EdgeNeXt、GhostNet、EfficientNet和LeViT。实验中使用的植物数据集为Plant Disease、Cassava和Turmeric。每个模型均使用准确率、F1分数、精确率和召回率指标进行评估。在三个数据集中，使用GhostNet架构在Turmeric数据集上取得了最佳性能，其得分分别为0.9978（准确率）、0.9976（F1分数）、0.9976（精确率）和0.9976（召回率）。",null,"Journal of Intelligent & Fuzzy Systems","2026-10-01T00:00:00Z","论文",10,false,70,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},12,20,16,13,9,1,"系统比较六种轻量网络在三种作物病害数据集上的精度效率权衡，对边缘端部署有参考价值，但属常规模型对比研究，公共影响有限。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","边缘计算","植物病害识别","轻量模型",{"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},[45,51,58],{"code":33,"score":13,"supported":46,"title":47,"abstract":49,"tags":50},true,[48],"deep learning",[],[28,30],{"code":52,"score":53,"supported":46,"title":54,"abstract":55,"tags":57},"I1",7,[],[27,56],"smart agriculture",[27],{"code":59,"score":22,"supported":14,"title":60,"abstract":61,"tags":62},"I5",[],[],[29],[64,65],"GhostNet 姜黄病害识别","轻量神经网络 植物病害分类","GhostNet姜黄病害识别-4009",6,"10.1177\u002F18758967261494075",{"doi":68,"openalex_id":70,"authors":71,"venue":10,"cited_by_count":75,"oa_url":9,"card":76,"direction":82,"ingested_from":83},"W7215069222",[72],{"name":73,"orcid":74},"Buket Toptaş","https:\u002F\u002Forcid.org\u002F0000-0003-2556-8199",0,{"tldr":77,"method":78,"finding":79,"direction":80,"opportunity":81},"比较六种轻量神经网络在三种植物病害数据集上的分类性能，面向边缘设备部署。","使用SqueezeNet、MobileViT、EdgeNeXt、GhostNet","GhostNet在Turmeric数据集上表现最佳，准确率达0.9978，适合资源受限设备。","农业人工智能与决策模型","可探索轻量模型在真实田间边缘设备上的实时病害检测与多作物泛化能力。","智慧农业 \u002F 农业物联网","openalex","2026-10-02T23:30:11.674439Z",{"total":67,"page":22,"page_size":67,"items":86},[87,151,228,283,364,414],{"id":88,"title":89,"url":90,"summary":91,"summary_zh":92,"content":9,"source_name":93,"source_url":90,"published_at":94,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":95,"score_detail":96,"sources":101,"tags":103,"primary_topic":106,"secondary_topics":112,"classification":114,"search_phrases":131,"slug":134,"view_count":75,"doi":135,"paper":136,"created_at":150},5592,"Evaluation of Concept-Rank Collapse Audits for Computer Vision Agriculture Disease Classifiers Under Low-Label Regimes: An Interpretability Capacity Bound for Smallholder Fungicide Selectivity","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545193","Concept-based interpretability methods for plant-disease classifiers are typically evaluated by whether their concept activations correlate with human-annotated labels, but this criterion is silent on a prior question: whether the classifier's internal representation retains enough concept-rank structure for any audit to be informative. the present analysis formalises that prior question as an interpretability capacity bound. A classifier is modelled as a concept-rank operator whose singular values govern how much concept-level signal survives the projection from pixel space to class logits, and a collapse audit is defined as the ratio of retained concept-rank energy to total concept energy. Under a low-label regime, where concept annotations are scarce and the audit must be estimated from few labelled examples, the bound predicts that audit reliability degrades as a function of the concept-rank spectrum and the label budget. The analysis is a theoretical and methodological contribution: it derives the bound, states the conditions under which a collapse audit is identifiable, and characterises the operating characteristics of the audit as a decision procedure under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 52 clusters were required for 5 percentage points at 80% power. The bound reframes concept audits as capacity tests rather than fidelity tests, with consequences for how smallholder fungicide-selectivity decisions should be supported by interpretable disease classifiers.","面向植物病害分类器的基于概念的可解释性方法，通常以其概念激活是否与人工标注标签相关作为评价标准，但该标准对一个更先前的问题保持沉默：分类器的内部表征是否保留了足够的概念秩结构，从而使任何审计具有信息量。本分析将该先前问题形式化为一个可解释性容量上界。分类器被建模为概念秩算子，其奇异值支配着从像素空间到类别logits的投影中概念级信号得以保留的程度，并将坍缩审计定义为保留的概念秩能量与总概念能量之比。在低标签情形下，概念标注稀缺且审计必须由少量标注样本估计，该上界预测审计可靠性会随概念秩谱和标签预算的函数而退化。本分析是一项理论和方法论贡献：推导出该上界，陈述坍缩审计可识别的条件，并在声明的方差结构（簇间SD = 6.0，残差SD = 4.5，组内相关系数 = 0.64）下刻画审计作为决策程序的操作特征。在2000个设计单元中，针对簇数为8、16、24和32以及效应量为3、5和8个百分点的情形估计了功效；在80%功效下达到5个百分点需要52个簇。该上界将概念审计重新框定为容量检验而非保真度检验，并对可解释病害分类器应如何支持小农杀菌剂选择性决策具有启示意义。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",69,{"impact":97,"substance":98,"depth":99,"authority":17,"freshness":13,"relevant":22,"comment":100},8,21,18,"理论方法论文，提出可解释性容量上界并给出审计功效分析，对智慧农业病害识别可信度有参考价值，但属小众方法学研究、缺乏田间实证，暂不宜进入每日精选。",[102],{"name":93,"url":90},[27,28,104,30,105],"小农户","可解释性",{"code":107,"slug":108,"name":109,"path":110,"parent_code":111},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[113],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":115},{"candidates":116,"agricultural_evidence":46},[117,122,127],{"code":107,"score":17,"supported":46,"title":118,"abstract":120,"tags":121},[119],"smallholder",[119],[104],{"code":33,"score":13,"supported":46,"title":123,"abstract":125,"tags":126},[124],"computer vision",[],[28,30],{"code":52,"score":22,"supported":14,"title":128,"abstract":129,"tags":130},[],[],[27],[132,133],"农业人工智能 植物病害识别 可解释性 智慧农业","农业人工智能 植物病害识别","农业人工智能植物病害识别可解释性智慧农业-5592","10.5281\u002Fzenodo.23545193",{"doi":135,"openalex_id":137,"authors":138,"venue":93,"cited_by_count":75,"oa_url":90,"card":145,"direction":80,"ingested_from":83},"W7223137323",[139,141,143],{"name":140,"orcid":9},"Akshay Ramgoolam",{"name":142,"orcid":9},"Priya Bundhoo Appadoo",{"name":144,"orcid":9},"Jean-Noël LEspérance",{"tldr":146,"method":147,"finding":148,"direction":80,"opportunity":149},"提出可解释性容量上界，评估低标签下概念秩崩溃审计的可靠性。","将分类器建模为概念秩算子，推导容量上界并做2000格功效模拟。","审计可靠性随概念秩谱与标签预算下降；5个百分点效应需52个聚类达80%功效。","可将容量上界用于小农户杀菌剂选择性决策的可解释模型审计与标签预算设计。","2026-10-22T11:21:10.425290Z",{"id":152,"title":153,"url":154,"summary":155,"summary_zh":156,"content":9,"source_name":157,"source_url":154,"published_at":158,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":159,"sources":162,"tags":164,"primary_topic":167,"secondary_topics":168,"classification":169,"search_phrases":203,"slug":206,"view_count":22,"doi":207,"paper":208,"created_at":227},5248,"An entropy-guided dynamic lightweight network for identification of wild edible vegetables of BTR","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02528-x","Wild edible vegetables of Bodoland Territorial Region (BTR) are the species of plants that grow naturally and are commonly consumed as food by the indigenous communities. These plants are important sources of nutritional and therapeutic properties and are essential for food security and socio-economic development of rural communities. Therefore, sustainable conservation and utilization of these unique plant resources are of great significance. Developing an efficient and accurate automatic identification system is an essential conservation strategy for them. Current plant identification methods are mainly based on classical machine learning (ML) methods and deep convolutional neural network (CNN) architectures for recognition and classification. These methods are effective for images taken under controlled or pristine conditions, but the performance is often far from ideal in the complex and natural environments of forest and grassland, where wild edible plants are found. This paper proposes an entropy-guided dynamic lightweight deep learning architecture for automatic detection of wild edible vegetables in BTR. The proposed model employs adaptive computation with Dynamic Kernel Selection (DKS), Region-wise Adaptive Feature Routing (RAFR), Ultra-Light Context Encoding (ULCE), and Entropy-Guided Global Pooling (EGGP) to capture multi-scale plant characteristics efficiently and reduce redundant computations. The novelty of our proposed framework lies in its coordinated integration of these modules that is designed to reduce computational overhead without compromising with performance. Experimental results show that the proposed model achieves excellent classification accuracy of 90.69% with significantly reduced parameter complexity and floating-point operations (FLOPs), which is well-suited for real-time precision agriculture and low-resource edge deployment applications.","博多兰领土区域（BTR）的野生食用蔬菜是自然生长并被土著社区普遍作为食物食用的植物物种。这些植物是营养和治疗特性的重要来源，对农村社区的粮食安全和社会经济发展至关重要。因此，对这些独特植物资源的可持续保护和利用具有重要意义。开发高效准确的自动识别系统是保护这些植物资源的重要策略。当前的植物识别方法主要基于经典机器学习（ML）方法和深度卷积神经网络（CNN）架构进行识别和分类。这些方法对于在受控或原始条件下拍摄的图像是有效的，但在野生食用植物所处的森林和草地复杂自然环境中，其性能往往远不理想。本文提出了一种熵引导的动态轻量级深度学习架构，用于自动检测BTR中的野生食用蔬菜。所提出的模型采用动态核选择（DKS）、区域自适应特征路由（RAFR）、超轻量上下文编码（ULCE）和熵引导全局池化（EGGP）的自适应计算，以高效捕捉多尺度植物特征并减少冗余计算。我们所提出框架的新颖性在于这些模块的协同集成，旨在在不影响性能的情况下降低计算开销。实验结果表明，所提出的模型达到了90.69%的优异分类准确率，同时显著降低了参数复杂度和浮点运算量（FLOPs），非常适合实时精准农业和低资源边缘部署应用。","Discover Artificial Intelligence","2026-10-17T00:00:00Z",{"impact":17,"substance":18,"depth":160,"authority":17,"freshness":21,"relevant":22,"comment":161},17,"提出熵引导动态轻量网络用于野生食用蔬菜识别，准确率90.69%且参数与算力显著降低，方法有新意、结论可靠，但属细分应用研究，公共影响有限。",[163],{"name":157,"url":154},[27,28,29,165,166],"野生蔬菜","植物识别",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":170},{"candidates":171,"agricultural_evidence":46},[172,179,185,192,199],{"code":33,"score":13,"supported":46,"title":173,"abstract":174,"tags":178},[],[175,176,177,48],"深度学习","机器学习","machine learning",[28],{"code":52,"score":53,"supported":46,"title":180,"abstract":181,"tags":184},[],[182,183],"精准农业","precision agriculture",[27],{"code":186,"score":67,"supported":46,"title":187,"abstract":188,"tags":191},"A8",[],[189,190],"粮食安全","food security",[],{"code":193,"score":194,"supported":46,"title":195,"abstract":196,"tags":198},"B1",3,[],[197],"蔬菜",[],{"code":59,"score":22,"supported":14,"title":200,"abstract":201,"tags":202},[],[],[29],[204,205],"BTR 野生食用蔬菜 识别","熵引导 轻量网络 植物识别","BTR野生食用蔬菜识别-5248","10.1007\u002Fs44163-026-02528-x",{"doi":207,"openalex_id":209,"authors":210,"venue":157,"cited_by_count":75,"oa_url":154,"card":222,"direction":80,"ingested_from":83},"W7222842686",[211,213,216,219],{"name":212,"orcid":9},"Tania Sarkar",{"name":214,"orcid":215},"Pranav Kumar Singh","https:\u002F\u002Forcid.org\u002F0000-0001-8987-0229",{"name":217,"orcid":218},"Parag Bhuyan","https:\u002F\u002Forcid.org\u002F0009-0006-4430-1851",{"name":220,"orcid":221},"Anshuman Kalla","https:\u002F\u002Forcid.org\u002F0000-0002-3985-4578",{"tldr":223,"method":224,"finding":225,"direction":80,"opportunity":226},"提出熵引导动态轻量网络，用于识别BTR野生食用蔬菜。","动态核选择、区域自适应特征路由、超轻上下文编码与熵引导池化。","分类准确率达90.69%，参数量和FLOPs显著降低，适合边缘部署。","可探索复杂自然场景下轻量模型在低资源边缘设备的实时植物识别与保护应用。","2026-10-18T23:30:58.805321Z",{"id":229,"title":230,"url":231,"summary":232,"summary_zh":233,"content":9,"source_name":234,"source_url":231,"published_at":235,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":236,"score_detail":237,"sources":239,"tags":241,"primary_topic":244,"secondary_topics":245,"classification":246,"search_phrases":263,"slug":266,"view_count":75,"doi":267,"paper":268,"created_at":282},5192,"A Comparative Study of Client-Level Stabilization Techniques in Federated Learning for Robust Plant Disease Recognition","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fs26206566","The successful integration of artificial intelligence (AI) algorithms in precision agriculture provided farmers with the means to identify potential issues, such as early disease detection or pest infestations, by making crop-related predictions. The new technological developments in this field improve the functioning of traditional agriculture by reducing the dependence on specialized human expertise and facilitating greater access to diagnostic procedures. However, most studies conducted in this area rely on centralized training of AI models, which leads to increased costs and raised privacy-related concerns. Federated learning (FL) offers a compelling framework for collaboratively training machine learning (ML) models on distributed datasets without centralizing the data owned by each participant. FL offers significant advantages in the preservation of data privacy, facilitates the collaborative exploitation of geographically distributed data sources, and reduces the risks associated with centralized collection. In this paper, we investigate the current limitations and performance boundaries of existing Federated and privacy-preserving learning in distributed sensing systems for plant disease detection, in scenarios that closely resemble real-world conditions. Our analysis specifically examined how FL aggregation algorithms, such as FedAvg, FedAvgM and FedProx, work in a custom multi-client experimental setup where data is partitioned so that each participant observes only those symptom patterns associated with specific environmental and climatic conditions. We propose further partial aggregation of the local model, an approach where we send only the model backbone to the central server, while the classification head of each participant remains strictly on the client side. The construction of environmentally heterogeneity is combined with a globally representative stabilizer node designed to preserve knowledge across disease classes that may be absent from certain clients.","人工智能（AI）算法在精准农业中的成功集成，为农民提供了通过作物相关预测来识别潜在问题（如早期病害检测或虫害侵染）的手段。该领域的新技术发展通过减少对专业人类知识的依赖并促进更广泛的诊断程序获取，改善了传统农业的运作方式。然而，该领域的大多数研究依赖于AI模型的集中式训练，这导致成本增加并引发隐私相关担忧。联邦学习（FL）提供了一个引人注目的框架，可在分布式数据集上协作训练机器学习（ML）模型，而无需集中化各参与者拥有的数据。联邦学习在保护数据隐私方面具有显著优势，促进了地理上分布的数据源的协作利用，并降低了与集中收集相关的风险。本文研究了在高度接近真实条件的场景中，现有联邦学习与隐私保护学习在用于植物病害检测的分布式感知系统中的当前局限性与性能边界。我们的分析特别考察了联邦学习聚合算法（如FedAvg、FedAvgM和FedProx）在自定义多客户端实验设置中的工作方式，其中数据被划分，使得每个参与者仅观察到与特定环境和气候条件相关的症状模式。我们提出对本地模型进行进一步的部分聚合，即仅将模型骨干发送至中央服务器，而每个参与者的分类头严格保留在客户端侧。环境异质性的构建与一个全局代表性的稳定器节点相结合，该节点旨在保留某些客户端可能缺失的跨病害类别的知识。","Sensors","2026-10-16T00:00:00Z",78,{"impact":99,"substance":98,"depth":160,"authority":20,"freshness":21,"relevant":22,"comment":238},"该研究针对植物病害识别中的联邦学习客户端稳定化方法提出部分聚合与稳定节点方案，方法新颖、实验贴近真实场景，对智慧农业隐私计算有参考价值，值得进入每日精选。",[240],{"name":234,"url":231},[27,28,242,30,243],"联邦学习","数据隐私",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":247},{"candidates":248,"agricultural_evidence":46},[249,254,258],{"code":33,"score":17,"supported":46,"title":250,"abstract":251,"tags":253},[],[176,177,252],"artificial intelligence",[28,30,242],{"code":52,"score":53,"supported":46,"title":255,"abstract":256,"tags":257},[],[182,183],[27],{"code":259,"score":22,"supported":14,"title":260,"abstract":261,"tags":262},"I6",[],[],[243],[264,265],"联邦学习 植物病害识别","FedProx 作物病害检测","联邦学习植物病害识别-5192","10.3390\u002Fs26206566",{"doi":267,"openalex_id":269,"authors":270,"venue":234,"cited_by_count":75,"oa_url":231,"card":277,"direction":80,"ingested_from":83},"W7222586734",[271,274],{"name":272,"orcid":273},"Aurora-Felicia Cristea","https:\u002F\u002Forcid.org\u002F0009-0003-9718-065X",{"name":275,"orcid":276},"Ciprian Dobre","https:\u002F\u002Forcid.org\u002F0000-0003-4638-7725",{"tldr":278,"method":279,"finding":280,"direction":80,"opportunity":281},"比较联邦学习客户端级稳定化技术，提升植物病害识别的鲁棒性。","FedAvg\u002FFedAvgM\u002FFedProx聚合，仅上传骨干网络、分类头留本地，","部分聚合加稳定节点可缓解环境异质性与类别缺失带来的性能退化。","可探索面向非独立同分布农业数据的个性化联邦学习与通信效率优化。","2026-10-17T23:30:45.845684Z",{"id":284,"title":285,"url":286,"summary":287,"summary_zh":288,"content":9,"source_name":289,"source_url":286,"published_at":235,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":95,"score_detail":290,"sources":292,"tags":294,"primary_topic":296,"secondary_topics":302,"classification":304,"search_phrases":324,"slug":327,"view_count":22,"doi":328,"paper":329,"created_at":363},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",{"impact":17,"substance":18,"depth":19,"authority":17,"freshness":21,"relevant":22,"comment":291},"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。",[293],{"name":289,"url":286},[27,28,29,175,295],"玉米病害",{"code":297,"slug":298,"name":299,"path":300,"parent_code":301},"A3","maize","玉米","\u002Ftopic\u002Ffield-crops\u002Fmaize","A",[303],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":305},{"candidates":306,"agricultural_evidence":46},[307,312,316,320],{"code":297,"score":308,"supported":46,"title":309,"abstract":310,"tags":311},14,[298],[299,298],[],{"code":33,"score":97,"supported":46,"title":313,"abstract":314,"tags":315},[],[175,48],[28,175],{"code":52,"score":53,"supported":46,"title":317,"abstract":318,"tags":319},[],[182,183],[27],{"code":59,"score":22,"supported":14,"title":321,"abstract":322,"tags":323},[],[],[29],[325,326],"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":328,"openalex_id":330,"authors":331,"venue":289,"cited_by_count":75,"oa_url":286,"card":357,"direction":362,"ingested_from":83},"W7222602850",[332,335,338,340,342,345,347,349,352,354],{"name":333,"orcid":334},"Awwab Mohammad","https:\u002F\u002Forcid.org\u002F0000-0002-1148-5678",{"name":336,"orcid":337},"Punith kumar","https:\u002F\u002Forcid.org\u002F0009-0005-5676-4194",{"name":339,"orcid":9},"Bhavani K",{"name":341,"orcid":9},"Sailee N. Panchbhai",{"name":343,"orcid":344},"Sagarkumar S. Badhiye","https:\u002F\u002Forcid.org\u002F0000-0002-0710-3761",{"name":346,"orcid":9},"Manoj Kumar S B",{"name":348,"orcid":9},"Dharamvir",{"name":350,"orcid":351},"Sumit Kumar Sar","https:\u002F\u002Forcid.org\u002F0000-0003-0384-9088",{"name":353,"orcid":9},"Rani Chhagan Dalvi",{"name":355,"orcid":356},"Yogesh Harishchandra Bhosale","https:\u002F\u002Forcid.org\u002F0000-0001-6901-1419",{"tldr":358,"method":359,"finding":360,"direction":80,"opportunity":361},"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。","农业遥感与作物表型","2026-10-17T23:30:28.248539Z",{"id":365,"title":366,"url":367,"summary":368,"summary_zh":369,"content":9,"source_name":370,"source_url":367,"published_at":371,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":372,"score_detail":373,"sources":375,"tags":377,"primary_topic":379,"secondary_topics":380,"classification":381,"search_phrases":398,"slug":401,"view_count":22,"doi":402,"paper":403,"created_at":413},4982,"A Nightshade-SWIN Transformer-Based Framework for Enhanced Multi-Crop Plant Disease Classification","https:\u002F\u002Fdoi.org\u002F10.17148\u002Fijarcce.2026.151037","Nightshade crops (Solanaceae) face significant yield threats from phenotypically similar foliar diseases that can be difficult to distinguish during the early stages of infection.Accurate and timely disease identification is therefore essential for reducing crop losses, improving disease management, and supporting sustainable agricultural practices.While Convolutional Neural Networks (CNNs) have demonstrated strong performance in image-based plant disease classification due to their ability to extract spatial and local features, they may have limitations in capturing long-range dependencies and global contextual relationships within leaf images.These limitations become particularly important when disease symptoms exhibit subtle differences in texture, color, shape, and lesion distribution or when images are captured against complex field backgrounds.To address these challenges, we propose Nightshade-SWIN, an optimized deep learning framework based on the Swin Transformer architecture.The proposed framework utilizes a hierarchical shifted-window self-attention mechanism to effectively capture both minute lesion textures and global leaf morphology.By progressively learning multi-scale representations, Nightshade-SWIN can identify discriminative pathological characteristics while maintaining computational efficiency.The model was evaluated on a curated dataset of 5,000 multicrop images collected under variable field conditions, including differences in illumination, background complexity, leaf orientation, and disease severity.Experimental results demonstrate that Nightshade-SWIN achieves a peak classification accuracy of 94.6%, outperforming ResNet50 and EfficientNet-B0 by 4.9% and 3.4%, respectively.These results indicate the effectiveness of transformer-based feature learning for distinguishing visually similar foliar diseases.Furthermore, interpretability analysis using LIME and Grad-CAM confirms that the model primarily focuses on biologically relevant pathological regions, such as lesions, discoloration, spots, and infected leaf areas.This enhances the transparency and reliability of the proposed framework and supports its potential application in precision agriculture, automated crop monitoring, and early disease diagnosis.","茄科作物（Solanaceae）面临表型相似叶部病害的严重产量威胁，这些病害在感染早期难以区分。因此，准确及时地识别病害对于减少作物损失、改善病害管理以及支持可持续农业实践至关重要。尽管卷积神经网络（CNN）因其提取空间和局部特征的能力而在基于图像的植物病害分类中表现出色，但它们在捕捉叶片图像中的长距离依赖关系和全局上下文关联方面可能存在局限。当病害症状在纹理、颜色、形状和病斑分布上表现出细微差异，或图像在复杂田间背景下拍摄时，这些局限尤为突出。为应对这些挑战，我们提出了Nightshade-SWIN，一种基于Swin Transformer架构的优化深度学习框架。该框架利用分层移位窗口自注意力机制，有效捕捉微小的病斑纹理和全局叶片形态。通过逐步学习多尺度表征，Nightshade-SWIN能够在保持计算效率的同时识别具有判别力的病理特征。该模型在包含5,000张多作物图像的数据集上进行了评估，这些图像采集自多变田间条件，包括光照、背景复杂度、叶片朝向和病害严重程度的差异。实验结果表明，Nightshade-SWIN达到了94.6%的最高分类准确率，分别优于ResNet50和EfficientNet-B0达4.9%和3.4%。这些结果表明了基于Transformer的特征学习在区分视觉相似叶部病害方面的有效性。此外，使用LIME和Grad-CAM进行的可解释性分析证实，该模型主要关注生物学相关的病理区域，如病斑、变色、斑点及受感染叶片区域。这增强了所提框架的透明度和可靠性，并支持其在精准农业、自动化作物监测和早期病害诊断中的潜在应用。","IJARCCE","2026-10-14T00:00:00Z",68,{"impact":17,"substance":18,"depth":160,"authority":13,"freshness":21,"relevant":22,"comment":374},"该论文提出基于Swin Transformer的茄科作物病害分类框架，在5000张多作物图像上达到94.6%准确率，方法新颖、数据规模可观，对智慧农业病害识别有参考价值，但属细分领域学术进展，公共影响有限。",[376],{"name":370,"url":367},[27,28,182,30,378],"茄科作物",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":382},{"candidates":383,"agricultural_evidence":46},[384,388,392],{"code":52,"score":97,"supported":46,"title":385,"abstract":386,"tags":387},[],[182,183],[27,182],{"code":33,"score":97,"supported":46,"title":389,"abstract":390,"tags":391},[],[175,48],[28,30],{"code":393,"score":194,"supported":46,"title":394,"abstract":395,"tags":397},"H8",[],[396],"可持续农业",[],[399,400],"Nightshade-SWIN 病害分类","Swin Transformer 植物病害","Nightshade-SWIN病害分类-4982","10.17148\u002Fijarcce.2026.151037",{"doi":402,"openalex_id":404,"authors":405,"venue":370,"cited_by_count":75,"oa_url":367,"card":408,"direction":82,"ingested_from":83},"W7221746668",[406],{"name":407,"orcid":9},"Dr. Barkha Joshi",{"tldr":409,"method":410,"finding":411,"direction":80,"opportunity":412},"提出Nightshade-SWIN框架，用Swin Transformer提升茄科多作物叶片病害分类","基于Swin Transformer的移位窗口自注意力，5000张田间多作物图像","分类准确率达94.6%，优于ResNet50和EfficientNet-B0，且模型关注病斑等生物相","可探索轻量化Transformer在移动端实时病害诊断及跨作物迁移学习，结合多模态数据提升田间泛化能力。","2026-10-15T23:30:08.177373Z",{"id":415,"title":416,"url":417,"summary":418,"summary_zh":9,"content":9,"source_name":419,"source_url":9,"published_at":420,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":95,"score_detail":421,"sources":424,"tags":426,"primary_topic":429,"secondary_topics":430,"classification":431,"search_phrases":453,"slug":456,"view_count":75,"doi":9,"paper":457,"created_at":465},4967,"SADR-VSS:一种用于轻量级农业目标检测的场景自适应动态路由视觉状态空间模块","https:\u002F\u002Fwww.ebiotrade.com\u002Fnewsf\u002F2026-10\u002F20261013084746521.htm","本文提出场景自适应动态路由视觉状态空间模块SADR-VSS,用于复杂田间条件下的轻量级农业目标检测。该方法引入场景自适应路由机制,动态协调局部卷积表示和长距离状态空间建模,从而在异质农业场景中实现更灵活的特征提取。大量实验表明,SADR-VSS实现了改善的检测性能,验证了其在嵌入式设备和农业自动化平台上的部署潜力。研究面向自动除草平台的高分辨率图像,在YOLOv11n基线基础上展示更集中的目标激活响应。","生物通 2026 年 10 月 13 日","2026-10-13T00:00:00Z",{"impact":422,"substance":18,"depth":160,"authority":97,"freshness":21,"relevant":22,"comment":423},15,"提出场景自适应动态路由视觉状态空间模块，在YOLOv11n基线上提升复杂田间轻量检测性能，方法新颖且面向嵌入式除草平台，具备较强专业参考价值。",[425],{"name":419,"url":417},[27,28,29,427,428],"农业目标检测","智能除草",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":432},{"candidates":433,"agricultural_evidence":46},[434,439,445,449],{"code":33,"score":17,"supported":46,"title":435,"abstract":437,"tags":438},[436],"目标检测",[436],[28],{"code":440,"score":194,"supported":46,"title":441,"abstract":442,"tags":444},"F4",[],[443],"除草",[],{"code":52,"score":22,"supported":14,"title":446,"abstract":447,"tags":448},[],[],[27],{"code":59,"score":22,"supported":14,"title":450,"abstract":451,"tags":452},[],[],[29],[454,455],"SADR-VSS 农业目标检测","YOLOv11n 自动除草","SADR-VSS农业目标检测-4967",{"doi":9,"openalex_id":9,"authors":458,"venue":9,"cited_by_count":75,"oa_url":9,"card":459,"direction":80,"ingested_from":464},[],{"tldr":460,"method":461,"finding":462,"direction":80,"opportunity":463},"提出场景自适应动态路由视觉状态空间模块SADR-VSS，用于复杂田间轻量级农业目标检测。","结合局部卷积与长距离状态空间建模，基于YOLOv11n在自动除草高分辨率图像上验","SADR-VSS提升检测性能，目标激活更集中，具备嵌入式农业平台部署潜力。","可探索动态路由机制在更多作物与杂草场景的泛化性，及与边缘计算硬件的协同优化。","agent","2026-10-15T00:06:08.643930Z"]