[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3274":3,"related-3274":47},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":6,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":23,"tags":25,"search_phrases":31,"slug":34,"view_count":35,"doi":36,"paper":37,"created_at":46},3274,"Farmer-Friendly Decision Support System Using Explainable Orangutan Artificial Intelligence for Soil Fertility","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs42979-026-05329-2","Farmer-Friendly Decision Support System Using Explainable Orangutan Artificial Intelligence for Soil Fertility。SN Computer Science",null,"SN Computer Science","2026-09-22T00:00:00Z","论文",10,false,68,{"impact":16,"substance":17,"depth":18,"authority":19,"freshness":20,"relevant":21,"comment":22},12,18,16,13,9,1,"论文提出可解释猩猩优化算法驱动的土壤肥力决策支持系统，方法新颖且面向农户，但尚属学术探索阶段，产业影响有限。",[24],{"name":9,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","可解释AI","决策支持系统","土壤肥力",[32,33],"农业人工智能 决策支持系统 土壤肥力 智慧农业","农业人工智能 决策支持系统","农业人工智能决策支持系统土壤肥力智慧农业-3274",0,"10.1007\u002Fs42979-026-05329-2",{"doi":36,"openalex_id":38,"authors":39,"venue":9,"cited_by_count":35,"oa_url":8,"card":8,"direction":44,"ingested_from":45},"W7214018467",[40,42],{"name":41,"orcid":8},"K. Komala Devi",{"name":43,"orcid":8},"Josephine Prem Kumar","智慧农业 \u002F 农业物联网","openalex","2026-09-23T23:30:12.314548Z",{"total":48,"page":21,"page_size":48,"items":49},6,[50,94,127,168,215,273],{"id":51,"title":52,"url":53,"summary":54,"summary_zh":55,"content":8,"source_name":56,"source_url":53,"published_at":57,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":58,"score_detail":59,"sources":63,"tags":65,"search_phrases":68,"slug":71,"view_count":35,"doi":72,"paper":73,"created_at":93},3368,"A PCA-based deep feature optimization framework for explainable orange fruit disease classification","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs12870-026-09984-8","Accurate classification of orange fruit diseases is important for precision agriculture and yield protection. This study develops and rigorously benchmarks a hybrid deep-feature framework for classifying Black Spot, Canker, Fresh, and Greening oranges (1,090 images), combining deep feature extraction, PCA-based dimensionality reduction, and classical machine-learning classification. Eight backbones (seven CNNs and a Vision Transformer, ViT) and four classifiers (32 configurations in total) were evaluated under 5 × 5 repeated stratified cross-validation, with PCA fitted exclusively on training-fold features in every iteration to eliminate data leakage. The proposed ViT + PCA+SVM configuration achieved the highest mean accuracy, 99.12%±0.71%, significantly outperforming every CNN-based backbone, including DenseNet201 + PCA + SVM (98.48%±0.81%, p \u003C 0.001). A dedicated variance-retention sensitivity analysis justifies the 98% threshold used throughout, and ablation experiments confirm that PCA substantially reduces feature dimensionality (by ~ 55.7% for ViT and ~ 76.6% for DenseNet201) without a significant loss of accuracy for either backbone. Explainability analysis — occlusion sensitivity and SHAP for the proposed ViT model, and Grad-CAM and SHAP for the DenseNet201 comparison model — shows that both configurations base predictions on biologically relevant, disease-affected regions of the fruit rather than spurious cues. These results identify ViT + PCA+SVM as the most accurate configuration evaluated, with DenseNet201 + PCA + SVM as a closely competitive, more compact convolutional alternative for intelligent orchard disease-monitoring systems.","橙类果实病害的准确分类对精准农业和产量保护具有重要意义。本研究开发并严格基准测试了一种混合深度特征框架，用于对黑斑病、溃疡病、新鲜和黄龙病橙类（1，090张图像）进行分类，该框架结合了深度特征提取、基于PCA的降维和经典机器学习分类。在5×5重复分层交叉验证下评估了八种骨干网络（七种CNN和一种视觉Transformer，ViT）和四种分类器（共32种配置），每次迭代中PCA仅在训练折特征上拟合以消除数据泄漏。所提出的ViT + PCA+SVM配置取得了最高平均准确率，为99.12%±0.71%，显著优于所有基于CNN的骨干网络，包括DenseNet201 + PCA + SVM（98.48%±0.81%，p \u003C 0.001）。专门的方差保留敏感性分析证明了全程使用的98%阈值是合理的，消融实验证实PCA大幅降低了特征维度（ViT约降低55.7%，DenseNet201约降低76.6%），且两种骨干网络均无显著准确率损失。可解释性分析——对所提出的ViT模型采用遮挡敏感性和SHAP，对DenseNet201对比模型采用Grad-CAM和SHAP——表明两种配置均基于果实中生物学相关的病害影响区域而非虚假线索进行预测。这些结果确定ViT + PCA+SVM为所评估的最准确配置，而DenseNet201 + PCA + SVM则是一种竞争力接近且更紧凑的卷积替代方案，可用于智能果园病害监测系统。","BMC Plant Biology","2026-09-23T00:00:00Z",79,{"impact":18,"substance":60,"depth":17,"authority":61,"freshness":20,"relevant":21,"comment":62},22,14,"方法严谨、数据与消融实验充分，对果园智能病害监测有实用参考价值，但属细分技术论文，产业影响有限。",[64],{"name":56,"url":53},[26,27,28,66,67],"病害识别","柑橘种植",[69,70],"柑橘病害 深度学习 分类","ViT PCA SVM 病害识别","柑橘病害深度学习分类-3368","10.1186\u002Fs12870-026-09984-8",{"doi":72,"openalex_id":74,"authors":75,"venue":56,"cited_by_count":35,"oa_url":53,"card":87,"direction":91,"ingested_from":45},"W7214068709",[76,78,80,82,85],{"name":77,"orcid":8},"Amruta Hingmire",{"name":79,"orcid":8},"Avinash Golande",{"name":81,"orcid":8},"Vinodkumar Bhutnal",{"name":83,"orcid":84},"Sagar Dhanraj Pande","https:\u002F\u002Forcid.org\u002F0000-0003-4506-6997",{"name":86,"orcid":8},"Tanuja Pande",{"tldr":88,"method":89,"finding":90,"direction":91,"opportunity":92},"提出PCA深度特征优化框架，用ViT+SVM分类橙子病害，准确率达99.12%。","8种骨干网络提取特征，PCA降维，4种分类器，5×5交叉验证。","ViT+PCA+SVM最优，PCA降维超55%且精度不降，可解释性验证有效。","农业人工智能与决策模型","可探索轻量化模型在移动端或边缘设备的实时病害检测与多作物泛化。","2026-09-24T23:30:34.116938Z",{"id":95,"title":96,"url":97,"summary":98,"summary_zh":99,"content":8,"source_name":100,"source_url":97,"published_at":101,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":102,"sources":106,"tags":108,"search_phrases":111,"slug":114,"view_count":35,"doi":115,"paper":116,"created_at":126},3197,"YieldVision: Model-Driven Decision Support System for Crop Yield Prediction","https:\u002F\u002Fdoi.org\u002F10.56201\u002Frjpst.vol.9.no1.2026.pg176.194","Agriculture, a cornerstone of global food security, faces unprecedented challenges in achieving sustainable productivity. Crop yield prediction, a crucial aspect of agricultural planning, suffers from inefficiencies rooted in technological and methodological gaps. While traditional approaches rely on historical data and empirical estimations, they often fail to address the complexity of environmental interactions because they lack real-time data integration and limited access to accurate data, leading to inaccuracies and resource mismanagement. Addressing these issues requires the development of advanced Machine Learning (ML) models-based system that can predict crop yield with high accuracy in dynamic environments. This paper presents YieldVision, a decision support system driven by advanced ML models for Crop Yield Prediction based on environmental, soil, and crop-specific factors. In this work, we proposed three crop prediction models: Random Forest (RF), Long Short Time Memory (LSTM) and Hybrid (RF LSTM). RF regressor effectively captured the complex, nonlinear relationships among soil nutrients, fertilizer application, and temperature. The LSTM network learned the way short-term weather fluctuations impact crop development over time, and the hybrid model combined LSTM for environmental-related factors, and RF for soil-related factors. Experimental results on the Crop Yield Prediction Dataset containing soil and weather parameters showed that RF has the highest accuracy 𝑅2 = 99.10% and outperforms both LSTM and hybrid which have 87.50% and 97.30% accuracy respectively, confirming its suitability for the dataset. RF has the lowest RMSE and MAE compared to LSTM and hybrid models. The models are integrated into a decision support system that is deployed to operationalize the models in real-world settings with user interfaces accessible via mobile or web-delivered real-time alerts and yield forecasts to farmers and agronomists.","农业作为全球粮食安全的基石，在实现可持续生产力方面面临着前所未有的挑战。作物产量预测作为农业规划的关键环节，因技术和方法上的不足而效率低下。传统方法依赖历史数据和经验估算，但由于缺乏实时数据集成且难以获取准确数据，往往无法应对环境相互作用的复杂性，导致预测不准确和资源管理不善。解决这些问题需要开发基于先进机器学习（ML）模型的系统，能够在动态环境中高精度地预测作物产量。本文提出了YieldVision，一个由先进机器学习模型驱动的决策支持系统，用于基于环境、土壤和作物特定因素的作物产量预测。在本研究中，我们提出了三种作物预测模型：随机森林（RF）、长短期记忆网络（LSTM）和混合模型（RF-LSTM）。随机森林回归器有效捕捉了土壤养分、施肥量和温度之间复杂的非线性关系。LSTM网络学习了短期天气波动如何随时间影响作物发育，而混合模型将LSTM用于环境相关因素，RF用于土壤相关因素。在包含土壤和天气参数的作物产量预测数据集上的实验结果表明，随机森林具有最高的准确率𝑅2 = 99.10%，优于LSTM和混合模型，后两者的准确率分别为87.50%和97.30%，证实了随机森林对该数据集的适用性。与LSTM和混合模型相比，随机森林具有最低的RMSE和MAE。这些模型被集成到一个决策支持系统中，该系统已部署用于在实际环境中运行这些模型，其用户界面可通过移动端或网页访问，向农民和农艺师提供实时警报和产量预测。","RESEARCH JOURNAL OF PURE SCIENCE AND TECHNOLOGY","2026-09-21T00:00:00Z",{"impact":16,"substance":103,"depth":18,"authority":16,"freshness":104,"relevant":21,"comment":105},20,8,"提出RF、LSTM及混合模型并集成决策支持系统，方法对比与精度数据扎实，对智慧农业产量预测有参考价值，但属单篇论文且来源期刊影响力有限。",[107],{"name":100,"url":97},[26,27,109,110,29],"产量预测","机器学习",[112,113],"YieldVision 作物产量预测","RF LSTM 混合模型 产量预测","YieldVision作物产量预测-3197","10.56201\u002Frjpst.vol.9.no1.2026.pg176.194",{"doi":115,"openalex_id":117,"authors":118,"venue":100,"cited_by_count":35,"oa_url":8,"card":121,"direction":91,"ingested_from":45},"W7213883348",[119],{"name":120,"orcid":8},"D.J.S. Sako",{"tldr":122,"method":123,"finding":124,"direction":91,"opportunity":125},"提出YieldVision决策支持系统，用RF、LSTM及混合模型预测作物产量并部署应用。","基于土壤、环境与作物数据，构建RF、LSTM和RF-LSTM混合模型并集成到We","随机森林表现最佳，R²达99.10%，优于LSTM（87.50%）和混合模型（97.30%）。","可探索多源实时数据融合与模型可解释性，提升跨区域泛化能力并降低对高精度历史数据的依赖。","2026-09-22T23:30:43.591303Z",{"id":128,"title":129,"url":130,"summary":131,"summary_zh":132,"content":8,"source_name":133,"source_url":130,"published_at":101,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":134,"score_detail":135,"sources":137,"tags":139,"search_phrases":142,"slug":145,"view_count":35,"doi":146,"paper":147,"created_at":167},3159,"A multi-agent multi-modal LLM with Monte Carlo tree search for interpretable plant disease classification","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffpls.2026.1937679","In plant disease classification tasks, existing task specific models for plant disease classification typically achieve high accuracy by employing tailored network architectures. However, these models are unable to provide reasoning chains, which reduces interpretability and undermines trustworthiness in practical applications. In addition, although multimodal large language models (MLLMs) have made significant progress in smart agriculture, their disease classification accuracy lags behind that of task-specific models due to insufficient domain knowledge. While fine-tuning can enhance domain-specific knowledge, relying solely on unimodal data or simple multimodal information remains insufficient for adequately improving classification accuracy. Specifically, the simple multimodal information in this context refers to a single global description of the entire image produced by an MLLM. To address these issues, this paper proposes a multi-agent collaborative classification framework that integrates Monte Carlo Tree Search (MCTS) with multimodal enhancement. With an MLLM as the core, the framework first introduces an MCTSguided multi-agent collaboration mechanism that directs the agents to progressively generate region-specific descriptive textual information, which together with the image forms complex multimodal information and thereby boosts classification performance. Meanwhile, the search trajectories generated by MCTS are leveraged to construct reasoning processes for enhanced interpretability. Finally, a self-training mechanism is adopted to iteratively optimize the model, further improving overall performance. Experimental results show that the framework improves the baseline MLLM’s accuracy while providing interpretable reasoning chains, achieving gains from 0.6701 to 0.9965 on 3-class potato, 0.5340 to 0.9651 on 5-class potato, 0.5387 to 0.9467 on tomato, and 0.4705 to 0.9948 on grape.","在植物病害分类任务中，现有的植物病害分类专用模型通常通过采用定制化的网络架构来实现高准确率。然而，这些模型无法提供推理链，这降低了可解释性，并在实际应用中削弱了可信度。此外，尽管多模态大语言模型（MLLMs）在智慧农业领域取得了显著进展，但由于领域知识不足，其病害分类准确率仍落后于专用模型。虽然微调可以增强领域特定知识，但仅依赖单模态数据或简单的多模态信息仍不足以充分提升分类准确率。具体而言，此处的简单多模态信息是指MLLM对整个图像生成的单一全局描述。为解决上述问题，本文提出了一种融合蒙特卡洛树搜索（MCTS）与多模态增强的多智能体协同分类框架。该框架以MLLM为核心，首先引入MCTS引导的多智能体协同机制，引导各智能体逐步生成区域特定的描述性文本信息，这些信息与图像共同构成复杂的多模态信息，从而提升分类性能。同时，利用MCTS生成的搜索轨迹构建推理过程，以增强可解释性。最后，采用自训练机制迭代优化模型，进一步提升整体性能。实验结果表明，该框架在提升基线MLLM准确率的同时提供了可解释的推理链，在三分类马铃薯上从0.6701提升至0.9965，在五分类马铃薯上从0.5340提升至0.9651，在番茄上从0.5387提升至0.9467，在葡萄上从0.4705提升至0.9948。","Frontiers in Plant Science",80,{"impact":17,"substance":60,"depth":17,"authority":61,"freshness":104,"relevant":21,"comment":136},"提出多智能体+MCTS+多模态增强的可解释植物病害分类框架，准确率大幅提升，方法新颖且结论可靠，对智慧农业AI应用有参考价值。",[138],{"name":133,"url":130},[26,27,28,140,141],"植物病害识别","多模态大模型",[143,144],"多智能体 蒙特卡洛树搜索 植物病害分类","多模态大模型 马铃薯 番茄 葡萄 病害识别","多智能体蒙特卡洛树搜索植物病害分类-3159","10.3389\u002Ffpls.2026.1937679",{"doi":146,"openalex_id":148,"authors":149,"venue":133,"cited_by_count":35,"oa_url":130,"card":162,"direction":44,"ingested_from":45},"W7213924931",[150,153,155,157,160],{"name":151,"orcid":152},"Ridong Wu","https:\u002F\u002Forcid.org\u002F0000-0003-0729-3099",{"name":154,"orcid":8},"Hua Zou",{"name":156,"orcid":8},"Zhijie Li",{"name":158,"orcid":159},"Qian Zhou","https:\u002F\u002Forcid.org\u002F0000-0001-7964-8130",{"name":161,"orcid":8},"一子 江﨑",{"tldr":163,"method":164,"finding":165,"direction":91,"opportunity":166},"提出多智能体多模态LLM结合蒙特卡洛树搜索，实现可解释的植物病害分类。","MCTS引导多智能体生成区域描述文本，构建复杂多模态信息并自训练优化。","分类准确率大幅提升，如马铃薯三分类从0.6701升至0.9965，并提供推理链。","可探索将MCTS多智能体框架迁移至其他作物病害或田间复杂场景，提升可解释性与泛化性。","2026-09-22T23:30:11.045394Z",{"id":169,"title":170,"url":171,"summary":172,"summary_zh":173,"content":8,"source_name":174,"source_url":171,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":175,"score_detail":176,"sources":179,"tags":181,"search_phrases":183,"slug":186,"view_count":35,"doi":187,"paper":188,"created_at":214},3153,"Explainable optimized deep learning and generative AI based framework for finger millet disease detection in smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02222-y","Abstract Eleusine coracana, locally known as finger millet (ragi), is a wholesome, climate-resilient crop. It is a staple food in semi-dry and dry areas of Asia and Africa. The productivity of this crop can be severely reduced by diseases such as downy, mottle, seedling, smut, and wilt. Conventional diagnosis methods are often time-consuming and unsuitable for large-scale farming. Therefore,Deep Learning (DL) offers an alternative for smart commercial farming that can streamline disease screening and improve productivity. This work offers an AI-integrated framework for early disease intervention combining DL, Explainable AI (XAI), and an image-based decision-support interface. The finger millet (ragi) dataset from Kaggle was used to train and evaluate custom CNN, VGG16, and ResNet50 models. To reduce feature redundancy and improve generalization, Grey Wolf Optimizer (GWO) based feature selection was applied to the penultimate-layer features of each model. Custom CNN, VGG16, and ResNet50 achieved accuracies of 89.40%, 83.11%, and 91.83%, respectively. Further, the use of GWO improved the accuracies to 97.71%, 90.84%, and 98.36%, respectively. ResNet50 achieved the highest accuracy 98.36% and was selected as the final backbone. The final ResNet50 model is integrated with the Gemini API to provide disease-specific recommendations based on the predicted class and user queries. To increase prediction transparency, Grad-CAM was used to highlight image regions that influenced the model output. The integration of DL, XAI, and generative AI provides a scalable approach for detecting and managing finger millet diseases. The proposed work boosts precision agriculture by providing AI-based and XAI-supported decision-making, while fostering sustainable agricultural practices.","摘要 穇子（Eleusine coracana），当地称为指黍（ragi），是一种有益健康且气候适应性强的作物。它是亚洲和非洲半干旱及干旱地区的主食。霜霉病、斑驳病、苗枯病、黑穗病和枯萎病等病害可严重降低该作物的产量。传统诊断方法往往耗时且不适合大规模种植。因此，深度学习（DL）为智能商业农业提供了一种替代方案，可简化病害筛查并提高生产力。本研究提出了一种人工智能集成框架，用于早期病害干预，结合了深度学习、可解释人工智能（XAI）和基于图像的决策支持界面。使用来自Kaggle的指黍（ragi）数据集训练和评估了自定义CNN、VGG16和ResNet50模型。为减少特征冗余并提高泛化能力，将灰狼优化器（GWO）基于特征选择应用于每个模型的倒数第二层特征。自定义CNN、VGG16和ResNet50分别达到了89.40%、83.11%和91.83%的准确率。此外，使用GWO将准确率分别提高到97.71%、90.84%和98.36%。ResNet50达到了最高准确率98.36%，并被选为最终骨干网络。最终的ResNet50模型与Gemini API集成，根据预测类别和用户查询提供针对特定病害的建议。为提高预测透明度，使用Grad-CAM突出显示影响模型输出的图像区域。深度学习、可解释人工智能和生成式人工智能的集成提供了一种可扩展的方法，用于检测和管理指黍病害。所提出的工作通过提供基于人工智能和可解释人工智能支持的决策，促进了精准农业，同时推动了可持续农业实践。","Discover Artificial Intelligence",75,{"impact":177,"substance":103,"depth":17,"authority":16,"freshness":12,"relevant":21,"comment":178},15,"将深度学习、可解释AI与生成式AI结合用于指状粟病害识别，方法新颖、数据充分，对智慧农业有参考价值。",[180],{"name":174,"url":171},[26,27,28,66,182],"小米作物",[184,185],"ResNet50 GWO 病害识别","农业人工智能 小米作物 智慧农业 病害识别","ResNet50GWO病害识别-3153","10.1007\u002Fs44163-026-02222-y",{"doi":187,"openalex_id":189,"authors":190,"venue":174,"cited_by_count":35,"oa_url":171,"card":209,"direction":44,"ingested_from":45},"W7213978567",[191,194,196,198,200,203,206],{"name":192,"orcid":193},"Sunil Kumar Mohapatra","https:\u002F\u002Forcid.org\u002F0000-0002-5865-095X",{"name":195,"orcid":8},"A. Sanjib Kumar Patro",{"name":197,"orcid":8},"Lulen Kumar Sahu",{"name":199,"orcid":8},"Chinmaye Dora",{"name":201,"orcid":202},"Sujata Chakravarty","https:\u002F\u002Forcid.org\u002F0000-0002-1293-5378",{"name":204,"orcid":205},"Kshira Sagar Sahoo","https:\u002F\u002Forcid.org\u002F0000-0002-6435-5738",{"name":207,"orcid":208},"Byomakesh Mahapatra","https:\u002F\u002Forcid.org\u002F0000-0002-9126-1729",{"tldr":210,"method":211,"finding":212,"direction":91,"opportunity":213},"提出融合深度学习、可解释AI与生成式AI的指状粟病害检测框架，实现高精度识别与决策支持。","用Kaggle指状粟图像训练CNN、VGG16、ResNet50，并用灰狼优化做","GWO特征选择显著提升精度，ResNet50达98.36%，结合Grad-CAM与Gemini AP","可探索轻量化模型与多作物泛化，并将XAI与生成式建议在田间移动端实时验证。","2026-09-22T23:30:10.255790Z",{"id":216,"title":217,"url":218,"summary":219,"summary_zh":220,"content":8,"source_name":221,"source_url":218,"published_at":101,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":58,"score_detail":222,"sources":224,"tags":226,"search_phrases":229,"slug":232,"view_count":35,"doi":233,"paper":234,"created_at":272},3137,"From patterns to prediction: detecting crop sequence and interpreting crop selection with explainable AI at field level in Denmark","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.agsy.2026.104982","CONTEXT Crop sequences are important for sustainable agriculture, yet we know relatively little about how they recur across individual fields or which information best predicts the crop grown each year. Denmark's national field records allow both questions to be examined over time. OBJECTIVE We identified recurrent crop sequences across Denmark and tested how well annual crop-family selection could be predicted from crop history, farm characteristics, management, prices, soil and climate. METHODS Using 10 years (2011−2020) of national field-level crop data (∼600,000 fields yr −1 ), we developed a heuristic algorithm to detect recurring crop sequence patterns. We then trained machine learning (LightGBM) and deep learning (TabNet) models to predict annual crop choices based on preceding crops and lagged management and farm structure predictors, and pedoclimatic conditions. Models were evaluated through forward-chaining validation and temporal holdout tests, and SHAP values were used to examine how LightGBM used each predictor. RESULTS AND DISCUSSION Crop sequences were largely dominated by cereals, with little diversification even in longer sequences. Crop selection patterns were strongly associated with the previous crops and farm typology, but pedoclimatic conditions and previous management played a minor role. The DL model achieved a slightly higher overall accuracy, particularly for dominant crops, while the ML model provided a more balanced performance across different crops and enabled interpretation through SHAP. SIGNIFICANCE The analysis separates two distinct tasks: identifying multi-year crop sequences and predicting the crop family grown each year. The sequence analysis shows why diversification cannot be assessed from sequence length or crop counts alone. The predictive models could help generate locally plausible sequences for agri-environmental simulations, reducing reliance on standard rotations that poorly reflect observed field histories.","背景 作物序列对可持续农业至关重要，但我们对它们在单个田块中如何重复出现，以及哪些信息最能预测每年种植的作物知之甚少。丹麦的国家田块记录使这两个问题得以在长时间尺度上加以考察。 目标 我们识别了丹麦全国范围内重复出现的作物序列，并检验了年度作物科选择能在多大程度上由作物历史、农场特征、管理、价格、土壤和气候来预测。 方法 利用10年（2011−2020年）的国家田块级作物数据（每年约60万块田），我们开发了一种启发式算法来检测重复出现的作物序列模式。随后，我们训练了机器学习（LightGBM）和深度学习（TabNet）模型，基于前茬作物以及滞后的管理和农场结构预测因子及土壤气候条件来预测年度作物选择。模型通过前向链式验证和时间留出测试进行评估，并使用SHAP值考察LightGBM如何利用每个预测因子。 结果与讨论 作物序列在很大程度上以谷类作物为主，即使在较长的序列中也几乎没有多样化。作物选择模式与前茬作物和农场类型密切相关，而土壤气候条件和先前管理的作用较小。深度学习模型取得了略高的总体准确率，尤其是在优势作物上，而机器学习模型在不同作物之间提供了更均衡的表现，并可通过SHAP进行解释。 意义 该分析区分了两个不同的任务：识别多年作物序列和预测每年种植的作物科。序列分析表明，为什么不能仅凭序列长度或作物计数来评估多样化。预测模型有助于为农业环境模拟生成局部合理的序列，减少对不能很好反映观测田块历史的标准轮作方式的依赖。","Agricultural Systems",{"impact":18,"substance":60,"depth":17,"authority":61,"freshness":20,"relevant":21,"comment":223},"基于丹麦十年全国田块数据，用可解释AI预测作物选择，方法新颖、数据规模大，对农业信息化与轮作模拟有参考价值。",[225],{"name":221,"url":218},[26,27,28,227,228],"作物轮作","丹麦农业",[230,231],"丹麦 作物序列 机器学习","LightGBM 作物选择 预测","丹麦作物序列机器学习-3137","10.1016\u002Fj.agsy.2026.104982",{"doi":233,"openalex_id":235,"authors":236,"venue":221,"cited_by_count":35,"oa_url":218,"card":267,"direction":91,"ingested_from":45},"W7213919617",[237,239,242,245,247,249,251,253,256,258,261,264],{"name":238,"orcid":8},"João G. Serra",{"name":240,"orcid":241},"Edwin Haas","https:\u002F\u002Forcid.org\u002F0000-0003-0664-642X",{"name":243,"orcid":244},"Diego Ábalos","https:\u002F\u002Forcid.org\u002F0000-0002-4189-5563",{"name":246,"orcid":8},"Søren Kolind Hvid",{"name":248,"orcid":8},"Tommy Dalgaard",{"name":250,"orcid":8},"Lars Uldall-Jessen",{"name":252,"orcid":8},"Franca Giannini-Kurina",{"name":254,"orcid":255},"Meshach Ojo Aderele","https:\u002F\u002Forcid.org\u002F0009-0007-9381-6445",{"name":257,"orcid":8},"Bo Thiesson",{"name":259,"orcid":260},"Jørgen Eivind Olesen","https:\u002F\u002Forcid.org\u002F0000-0002-6639-1273",{"name":262,"orcid":263},"Klaus Butterbach‐Bahl","https:\u002F\u002Forcid.org\u002F0000-0001-9499-6598",{"name":265,"orcid":266},"Jaber Rahimi","https:\u002F\u002Forcid.org\u002F0000-0002-2754-2358",{"tldr":268,"method":269,"finding":270,"direction":91,"opportunity":271},"用丹麦十年田块数据识别轮作模式并预测年度作物选择。","60万田块年数据，启发式序列检测+LightGBM\u002FTabNet，SHAP解释。","作物序列以谷物为主、多样化低；作物选择主要由前茬和农场类型决定。","可结合可解释AI生成区域化可行轮作方案，替代脱离实际的标准化轮作假设。","2026-09-22T23:30:05.518307Z",{"id":274,"title":275,"url":276,"summary":277,"summary_zh":278,"content":8,"source_name":279,"source_url":276,"published_at":101,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":280,"score_detail":281,"sources":285,"tags":287,"search_phrases":290,"slug":293,"view_count":35,"doi":294,"paper":295,"created_at":325},3134,"DAC-Grad-CAM: Defect-aware explainable AI for automated mango quality classification","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112443","Deep learning models achieve high fruit classification accuracy but lack interpretability; standard Grad-CAM produces diffuse attention without defect-specific localization. Existing explainable AI (XAI) methods for fruit grading rely on qualitative visual inspection, offering no quantitative validation of defect alignment, thereby limiting trust and regulatory acceptance in commercial quality control. This study presents an explainable mango classification framework that integrates classification performance with quantitative interpretability analysis through Defect-Aware Contrastive Grad-CAM (DAC-Grad-CAM). A new method that multiplicatively fuses gradient-based attention with an unsupervised LAB color-space bruise prior. A two-stage pipeline combining YOLO-11n detection with four classification architectures (ResNet18, MobileNetV3, VGG16, and a custom CNN). Evaluated on a three-class mango dataset (premium, class 1, class 2). Transfer-learning architectures substantially outperformed the custom CNN (ResNet18: 97.8%, VGG16: 97.6%, MobileNetV3: 96.5% vs 75.4%), a 22–23 percentage-point gap, with premium-class precision > 0.995. DAC-Grad-CAM produced the most concentrated explanations (entropy 10.02 versus 10.27 for Grad-CAM and 10.18 for SHAP). Validated against 50 manually annotated pixel-level bruise masks, it improved peak localization over standard Grad-CAM (pointing-game accuracy 0.561 versus 0.439) and, on the transfer-learning backbones, exceeded SHAP (0.659 versus 0.615) while requiring 34 ms per region against 4220 ms, a 124-fold reduction in cost. The Defect Focus Score (DFS) enabled task-specific validation, and the agreement between gradient and perturbation-based explanations (r = 0.68–0.75, p \u003C 0.001) confirmed that the localized features are task-genuine rather than method-specific artifacts. Under controlled perturbations emulating varied lighting, resolution, defocus, compression, and occlusion, defect-focused attention remained stable while classification accuracy degraded, and end-to-end throughput reached 17.4 fruit s⁻ 1 . The framework supports a confidence-based human-AI workflow (margin > 0.95, DFS > 0.37, r > 0.60) for trustworthy automated quality control in post-harvest operations.","深度学习模型虽能实现高精度水果分类，却缺乏可解释性；标准Grad-CAM产生的注意力分布弥散，无法实现缺陷特异性定位。现有用于水果分级的可解释人工智能（XAI）方法依赖定性视觉检查，无法对缺陷对齐进行定量验证，从而限制了其在商业质量控制中的可信度与法规接受度。本研究提出了一种可解释芒果分类框架，通过缺陷感知对比Grad-CAM（DAC-Grad-CAM）将分类性能与定量可解释性分析相结合。该方法将基于梯度的注意力与无监督LAB色彩空间瘀伤先验进行乘法融合。采用两阶段流水线，将YOLO-11n检测与四种分类架构（ResNet18、MobileNetV3、VGG16及自定义CNN）相结合。在三类芒果数据集（特级、一级、二级）上进行评估。迁移学习架构显著优于自定义CNN（ResNet18：97.8%，VGG16：97.6%，MobileNetV3：96.5% vs 75.4%），差距达22–23个百分点，特级类精确率> 0.995。DAC-Grad-CAM产生了最集中的解释（熵10.02，对比Grad-CAM的10.27和SHAP的10.18）。针对50个手动标注的像素级瘀伤掩膜进行验证，其峰值定位优于标准Grad-CAM（指向游戏准确率0.561 vs 0.439），且在迁移学习骨干网络上超过SHAP（0.659 vs 0.615），同时每区域仅需34 ms，而SHAP需4220 ms，成本降低124倍。缺陷聚焦分数（DFS）实现了任务特异性验证，基于梯度与基于扰动的解释之间的一致性（r = 0.68–0.75，p \u003C 0.001）证实了所定位特征是任务真实的，而非方法特异性伪影。在模拟不同光照、分辨率、散焦、压缩和遮挡的受控扰动下，缺陷聚焦注意力保持稳定，而分类准确率下降，端到端吞吐量达到17.4果·秒⁻¹。该框架支持基于置信度的人机协作工作流（margin > 0.95，DFS > 0.37，r > 0.60），可用于采后作业中可信赖的自动化质量控制。","Computers and Electronics in Agriculture",81,{"impact":18,"substance":282,"depth":283,"authority":61,"freshness":20,"relevant":21,"comment":284},23,19,"方法新颖、量化验证扎实的芒果品质分级可解释AI研究，对采后智能质检有直接参考价值。",[286],{"name":279,"url":276},[26,27,28,288,289],"采后处理","水果分级",[291,292],"芒果 品质分级 深度学习","DAC-Grad-CAM 芒果 缺陷检测","芒果品质分级深度学习-3134","10.1016\u002Fj.compag.2026.112443",{"doi":294,"openalex_id":296,"authors":297,"venue":279,"cited_by_count":35,"oa_url":276,"card":320,"direction":91,"ingested_from":45},"W7213923596",[298,300,302,304,306,308,311,313,316,318],{"name":299,"orcid":8},"Arshed Ahmed",{"name":301,"orcid":8},"Zhao Zhang",{"name":303,"orcid":8},"Ming Li",{"name":305,"orcid":8},"Shahram Hamza Manzoor",{"name":307,"orcid":8},"Zhanjiang Zhu",{"name":309,"orcid":310},"Rashed Ahmed","https:\u002F\u002Forcid.org\u002F0009-0000-7170-2200",{"name":312,"orcid":8},"Noor Gul",{"name":314,"orcid":315},"Mustafa Mhamed","https:\u002F\u002Forcid.org\u002F0000-0002-3106-669X",{"name":317,"orcid":8},"Bing Liu",{"name":319,"orcid":8},"Mahmoud A. Abdelhamid",{"tldr":321,"method":322,"finding":323,"direction":91,"opportunity":324},"提出DAC-Grad-CAM可解释框架，实现芒果缺陷感知分类与定量验证。","融合梯度注意力与LAB色空间瘀伤先验，YOLO-11n检测加四种分类网络。","ResNet18达97.8%精度，定位优于Grad-CAM与SHAP且快124倍。","可将缺陷感知XAI扩展至多水果多缺陷，并构建人机协同质控标准。","2026-09-22T23:30:02.407104Z"]