[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3487":3,"related-3487":67},{"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":18,"tags":20,"search_phrases":24,"slug":27,"view_count":15,"doi":28,"paper":29,"created_at":66},3487,"Enhanced YOLOv8-seg-SPDConv for accurate Schatzker classification of tibial plateau fractures","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffmed.2026.1843762","Background Tibial plateau fractures are typically caused by direct or indirect violent forces, often resulting in collapse fractures of the medial or lateral tibial plateau. Accurate diagnosis and classification of tibial plateau fractures are of significant importance for determining appropriate treatment plans. However, manual Schatzker classification is subject to considerable inter-observer variability, particularly among junior physicians and non-specialist clinicians, often resulting in suboptimal accuracy and prolonged decision-making time in high-throughput emergency settings. Objective This study aimed to develop and validate a lightweight YOLOv8n-seg-SPDConv model for the joint instance segmentation and Schatzker classification of tibial plateau fractures using thin-slice axial CT images from 552 unique patients. Specifically, by integrating the Space-to-Depth Convolution (SPDConv) module, we sought to preserve sub-2 mm cortical disruptions while reducing model complexity and suggesting potential for future deployment on resource-constrained platforms. Using a hold-out test set benchmarked against a senior orthopedic surgeon's consensus standard, we systematically evaluated the model's diagnostic performance, inference efficiency (226 FPS), and interpretability via Gradient-weighted Class Activation Mapping (Grad-CAM), with the ultimate goal of serving as a reliable real-time clinical adjunct that supports, rather than supplants, clinicians in emergency and perioperative decision-making. Methods From October 2017 to January 2026, a dataset comprising 552 axial CT slices from unique patients was collected. Each image was independently annotated into six clinically relevant categories according to the Schatzker classification. To prevent data leakage, dataset partitioning was strictly performed prior to augmentation using an 8:2 patient-wise split. The YOLOv8n-seg network was employed for joint segmentation and classification, with multiple preprocessing techniques applied to ensure data consistency and enhance model performance. Results The baseline YOLOv8n-seg achieved mAP50 values of 0.941, 0.956, 0.861, 0.818, 0.916, and 0.869 for Schatzker types I–VI, respectively. The improved YOLOv8n-seg-SPDConv yielded mAP50 of 0.928, 0.952, 0.929, 0.859, 0.925, and 0.893 for the six subtypes, with an overall mAP50 increasing from 0.893 to 0.914. Notably, the proposed model reduced parameters from 3.01 M to 2.64 M and FLOPs from 8.1 G to 7.3 G, while increasing inference speed from 188 to 226 FPS, meeting real-time requirements for edge deployment. Grad-CAM visualization confirmed that the model focused its attention on clinically meaningful regions, including fracture fissures and articular depressions, rather than background artifacts. Discussion This deep learning-based classification method provides efficient and reliable automated assessments to assist and augment manual evaluation by clinicians, with the potential to help reduce interobserver variability among junior physicians during emergency and perioperative decision-making. It shows promise as a tool that may help mitigate interobserver variability and support junior physicians in emergency and primary care settings. While marginal performance variations were observed in simple linear fractures, the model achieved substantial gains in complex, low-contrast subtypes. These findings highlight its value as a complementary clinical decision support tool, with the potential to support and streamline diagnostic workflows rather than replace expert judgment, ultimately contributing to improved patient outcomes.","背景 胫骨平台骨折通常由直接或间接暴力所致，常导致内侧或外侧胫骨平台塌陷骨折。胫骨平台骨折的准确诊断与分型对于确定合适的治疗方案具有重要意义。然而，人工Schatzker分型存在较大的观察者间差异，尤其在低年资医师和非专科临床医生中更为明显，常导致在高通量急诊环境下准确率欠佳且决策时间延长。目的 本研究旨在开发并验证一种轻量级YOLOv8n-seg-SPDConv模型，利用552例独特患者的薄层轴位CT图像，对胫骨平台骨折进行联合实例分割与Schatzker分型。具体而言，通过整合空间到深度卷积（SPDConv）模块，我们力求在保留亚2 mm皮质中断的同时降低模型复杂度，并为其未来在资源受限平台上的部署提供可能。使用以资深骨科外科医生共识标准为基准的留出测试集，我们系统评估了模型的诊断性能、推理效率（226 FPS）以及通过梯度加权类激活映射（Grad-CAM）实现的可解释性，最终目标是将其作为一种可靠的实时临床辅助工具，在急诊和围手术期决策中支持而非取代临床医生。方法 自2017年10月至2026年1月，收集了包含552例独特患者轴位CT切片的 dataset。每张图像均根据Schatzker分型独立标注为六个临床相关类别。为防止数据泄漏，数据集划分严格在增强之前进行，采用8：2的患者层面拆分。采用YOLOv8n-seg网络进行联合分割与分类，并应用多种预处理技术以确保数据一致性并提升模型性能。结果 基线YOLOv8n-seg对Schatzker I–VI型的mAP50分别为0.941、0.956、0.861、0.818、0.916和0.869。改进后的YOLOv8n-seg-SPDConv对六个亚型的mAP50分别为0.928、0.952、0.929、0.859、0.925和0.893，总体mAP50从0.893提升至0.914。值得注意的是，所提出的模型将参数量从3.01 M降至2.64 M，FLOPs从8.1 G降至7.3 G，同时增",null,"Frontiers in Medicine","2026-09-22T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"属医学影像AI研究，与三农、农业信息化、智慧农业主题无关，不予入选。",[19],{"name":10,"url":6},[21,22,23],"深度学习","医学影像","骨折分型",[25,26],"YOLOv8-seg-SPDConv 胫骨平台骨折","Schatzker 分型 深度学习","YOLOv8-seg-SPDConv胫骨平台骨折-3487","10.3389\u002Ffmed.2026.1843762",{"doi":28,"openalex_id":30,"authors":31,"venue":10,"cited_by_count":15,"oa_url":57,"card":58,"direction":64,"ingested_from":65},"W7213974124",[32,34,36,38,40,42,44,46,48,51,53,55],{"name":33,"orcid":9},"Haoteng Wang",{"name":35,"orcid":9},"Zhiheng Gao",{"name":37,"orcid":9},"Heting Xiao",{"name":39,"orcid":9},"Xuwei Ling",{"name":41,"orcid":9},"Haowen Lu",{"name":43,"orcid":9},"Yu Chen",{"name":45,"orcid":9},"Minbo Jian",{"name":47,"orcid":9},"Zheming Shen",{"name":49,"orcid":50},"Quan Zhou","https:\u002F\u002Forcid.org\u002F0000-0003-2376-7741",{"name":52,"orcid":9},"Peng Yang",{"name":54,"orcid":9},"Tao Liu",{"name":56,"orcid":9},"Yusen Qiao","https:\u002F\u002Fwww.frontiersin.org\u002Fjournals\u002Fmedicine\u002Farticles\u002F10.3389\u002Ffmed.2026.1843762\u002Fpdf",{"tldr":59,"method":60,"finding":61,"direction":62,"opportunity":63},"提出轻量YOLOv8n-seg-SPDConv模型，用CT图像实现胫骨平台骨折Schatzker分型","552例薄层轴位CT，YOLOv8n-seg加SPDConv模块，患者级8:2划","整体mAP50从0.893升至0.914，参数与FLOPs下降，推理达226 FPS。","农业人工智能与决策模型","该轻量分割-分类联合框架可迁移至农业图像细粒度分级任务，如作物病害或果实损伤分型。","农业遥感与作物表型","openalex","2026-09-25T23:30:22.979253Z",{"total":68,"page":69,"page_size":68,"items":70},6,1,[71,120,164,208,248,284],{"id":72,"title":73,"url":74,"summary":75,"summary_zh":76,"content":9,"source_name":77,"source_url":74,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":78,"score_detail":79,"sources":86,"tags":88,"search_phrases":93,"slug":96,"view_count":15,"doi":97,"paper":98,"created_at":119},3479,"Artificial Intelligence-Enabled Mangrove Ecosystem Monitoring Using Remote Sensing and Environmental Data","https:\u002F\u002Fdoi.org\u002F10.68012\u002Fair.v1i2.252","Mangrove ecosystems play a critical role in coastal protection, carbon sequestration, biodiversity conservation, and climate change mitigation; however, increasing anthropogenic pressures and environmental changes have accelerated mangrove degradation, creating an urgent need for efficient and scalable monitoring approaches. This study aims to develop an Artificial Intelligence-Enabled framework for monitoring mangrove ecosystem conditions by integrating remote sensing imagery with environmental datasets to improve the accuracy and timeliness of ecosystem assessment in tropical coastal regions. The proposed method combines multispectral satellite images, including vegetation indices derived from remote sensing data, with environmental variables such as temperature, precipitation, salinity, and tidal information, which are subsequently processed using a deep learning-based classification model to identify and categorize mangrove health conditions. Experimental evaluation demonstrates that the integration of remote sensing and environmental data significantly enhances model performance compared with approaches relying solely on satellite imagery, achieving high classification accuracy and improving the detection of early signs of ecosystem degradation. The findings further reveal that environmental parameters contribute substantially to distinguishing healthy, moderately degraded, and severely degraded mangrove areas across heterogeneous coastal environments. Consequently, the proposed framework provides an intelligent and reliable decision support tool for environmental monitoring agencies and policymakers while contributing to the development of resilient coastal ecosystem management and sustainable environmental governance in tropical archipelagic regions.","红树林生态系统在海岸防护、碳固存、生物多样性保护及气候变化减缓中发挥着关键作用；然而，日益加剧的人为压力和环境变化加速了红树林退化，亟需高效且可扩展的监测方法。本研究旨在开发一个人工智能驱动的框架，通过整合遥感影像与环境数据集来监测红树林生态系统状况，以提高热带沿海地区生态系统评估的准确性和时效性。所提出的方法将多光谱卫星影像（包括由遥感数据衍生的植被指数）与温度、降水、盐度和潮汐信息等环境变量相结合，随后使用基于深度学习的分类模型进行处理，以识别和分类红树林健康状况。实验评估表明，与仅依赖卫星影像的方法相比，遥感与环境数据的整合显著提升了模型性能，实现了较高的分类精度，并改善了对生态系统退化早期迹象的检测。研究结果进一步揭示，环境参数对于区分异质性沿海环境中健康、中度退化和严重退化的红树林区域具有重要贡献。因此，所提出的框架为环境监测机构和政策制定者提供了一种智能且可靠的决策支持工具，同时有助于热带群岛地区韧性沿海生态系统管理和可持续环境治理的发展。","AI Innovation and Resilience for the Environment (AIR)",76,{"impact":80,"substance":81,"depth":82,"authority":83,"freshness":84,"relevant":69,"comment":85},18,21,17,12,8,"将遥感与环境数据融合的深度学习框架用于红树林健康监测，方法有创新且结论可靠，对沿海生态治理有参考价值，但属细分领域研究，公共影响有限。",[87],{"name":77,"url":74},[89,21,90,91,92],"农业人工智能","遥感监测","生态监测","红树林",[94,95],"红树林 遥感 人工智能","红树林生态系统 监测","红树林遥感人工智能-3479","10.68012\u002Fair.v1i2.252",{"doi":97,"openalex_id":99,"authors":100,"venue":77,"cited_by_count":15,"oa_url":112,"card":113,"direction":118,"ingested_from":65},"W7213895268",[101,104,107,109],{"name":102,"orcid":103},"Dirvi Surya Abbas","https:\u002F\u002Forcid.org\u002F0000-0002-7819-2837",{"name":105,"orcid":106},"Asep Sutarman","https:\u002F\u002Forcid.org\u002F0009-0002-1029-7963",{"name":108,"orcid":9},"Ryan Davis",{"name":110,"orcid":111},"Maulana Abbas","https:\u002F\u002Forcid.org\u002F0009-0009-0137-9650","https:\u002F\u002Fjournal.sundarapublishing.com\u002Findex.php\u002Fair\u002Farticle\u002Fdownload\u002F252\u002F142",{"tldr":114,"method":115,"finding":116,"direction":64,"opportunity":117},"融合遥感影像与环境数据，用深度学习构建红树林生态系统健康监测框架。","多光谱卫星影像与植被指数，结合温度、降水、盐度、潮汐等环境变量，训练深度学习分类","融合环境数据显著提升分类精度，能更早识别红树林退化迹象，有效区分不同退化程度。","可探索多源时序遥感与环境数据融合的早期退化预警，并迁移至其他滨海湿地生态系统监测。","智慧农业 \u002F 农业物联网","2026-09-25T23:30:14.373467Z",{"id":121,"title":122,"url":123,"summary":124,"summary_zh":9,"content":9,"source_name":125,"source_url":123,"published_at":126,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":127,"score_detail":128,"sources":133,"tags":135,"search_phrases":138,"slug":141,"view_count":15,"doi":142,"paper":143,"created_at":163},3463,"Cotton leaf disease classification using deep learning models for smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs40066-026-00610-2","Cotton leaf disease classification using deep learning models for smart agriculture。Agriculture & Food Security","Agriculture & Food Security","2026-09-25T00:00:00Z",66,{"impact":83,"substance":129,"depth":130,"authority":131,"freshness":13,"relevant":69,"comment":132},16,15,13,"核心期刊论文，方法有一定参考价值，但属细分技术研究，产业影响有限。",[134],{"name":125,"url":123},[136,89,21,137],"智慧农业","棉花病害",[139,140],"棉花叶部病害 深度学习 分类","农业人工智能 智慧农业 棉花病害 深度学习","棉花叶部病害深度学习分类-3463","10.1186\u002Fs40066-026-00610-2",{"doi":142,"openalex_id":144,"authors":145,"venue":125,"cited_by_count":15,"oa_url":123,"card":9,"direction":118,"ingested_from":65},"W7214238516",[146,148,151,154,157,160],{"name":147,"orcid":9},"Hina Kiran Abbas",{"name":149,"orcid":150},"Muhammad Farrukh Shahid","https:\u002F\u002Forcid.org\u002F0009-0004-8787-1868",{"name":152,"orcid":153},"Rehab Bahaaddin Ashari","https:\u002F\u002Forcid.org\u002F0000-0003-1225-7535",{"name":155,"orcid":156},"Arwa Mashat","https:\u002F\u002Forcid.org\u002F0000-0002-0612-6005",{"name":158,"orcid":159},"Tariq Jamil Saifullah Khanzada","https:\u002F\u002Forcid.org\u002F0000-0003-1617-4403",{"name":161,"orcid":162},"M. Hassan Tanveer","https:\u002F\u002Forcid.org\u002F0000-0001-9266-6368","2026-09-25T23:30:09.012501Z",{"id":165,"title":166,"url":167,"summary":168,"summary_zh":169,"content":9,"source_name":170,"source_url":167,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":171,"score_detail":172,"sources":177,"tags":179,"search_phrases":182,"slug":185,"view_count":15,"doi":186,"paper":187,"created_at":207},3277,"Downscaling of SMAP Soil Moisture Based on the Transformer Algorithm in Anhui Province","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18193272","Soil moisture (SM) is critical for climate, water, and agriculture, but Soil Moisture Active Passive (SMAP) passive microwave products have coarse resolution, limiting regional applications. This study develops an SM downscaling framework based on Transformer and its variants (PatchTST and iTransformer), integrating multi-source satellite and groundwater data to generate 1 km daily SM products (2015–2022). Compared with Random Forest (RF), Long Short-Term Memory (LSTM), and Convolutional Neural Network–LSTM (CNN-LSTM), Transformer and its variants achieve superior accuracy and generalization. Validated against in situ measurements and SMCI1.0, the Transformer-downscaled SM product achieved the best accuracy with ubRMSE = 0.0372 m3\u002Fm3 and RMSE = 0.0591 m3\u002Fm3. The downscaled SM dataset not only captured finer spatial details but also preserved the spatial patterns and seasonal dynamics of the original SMAP product and showed good responsiveness to precipitation events. Feature importance analysis revealed that, aside from precipitation, the diurnal land surface temperature difference had a greater impact on SM than individual daytime or nighttime land surface temperature, ranking just below vegetation indices and soil texture factors, while groundwater level showed higher importance than elevation and surface temperature. This study confirms the effectiveness of Transformer-based models for SM spatial downscaling, providing a novel framework integrating remote sensing and deep hydrological information to generate accurate 1 km SM products.","土壤水分（SM）对气候、水资源和农业至关重要，但土壤水分主动被动（SMAP）被动微波产品分辨率较粗，限制了区域应用。本研究构建了一个基于Transformer及其变体（PatchTST和iTransformer）的土壤水分降尺度框架，融合多源卫星和地下水数据，生成1 km日尺度土壤水分产品（2015—2022年）。与随机森林（RF）、长短期记忆网络（LSTM）和卷积神经网络—长短期记忆网络（CNN-LSTM）相比，Transformer及其变体取得了更高的精度和泛化能力。利用站点实测数据和SMCI1.0进行验证，Transformer降尺度土壤水分产品精度最优，ubRMSE = 0.0372 m³\u002Fm³，RMSE = 0.0591 m³\u002Fm³。降尺度土壤水分数据集不仅捕捉到了更精细的空间细节，还保留了原始SMAP产品的空间格局和季节动态，并对降水事件表现出良好的响应。特征重要性分析表明，除降水外，昼夜地表温差对土壤水分的影响大于单独的白天或夜间地表温度，其重要性仅次于植被指数和土壤质地因子，而地下水埋深的重要性高于高程和地表温度。本研究证实了基于Transformer的模型在土壤水分空间降尺度中的有效性，为融合遥感和深层水文信息生成准确的1 km土壤水分产品提供了一种新框架。","Remote Sensing",81,{"impact":80,"substance":173,"depth":80,"authority":174,"freshness":175,"relevant":69,"comment":176},22,14,9,"基于Transformer的SMAP土壤水分1km降尺度研究，方法新颖、验证充分，对区域农业旱情监测有实用价值。",[178],{"name":170,"url":167},[136,89,21,180,181],"遥感","土壤墒情",[183,184],"SMAP 土壤水分 降尺度","Transformer 土壤水分 安徽","SMAP土壤水分降尺度-3277","10.3390\u002Frs18193272",{"doi":186,"openalex_id":188,"authors":189,"venue":170,"cited_by_count":15,"oa_url":167,"card":202,"direction":64,"ingested_from":65},"W7208807695",[190,192,194,196,198,200],{"name":191,"orcid":9},"Yuyang Fan",{"name":193,"orcid":9},"Jianwei Ma",{"name":195,"orcid":9},"Mengmeng Li",{"name":197,"orcid":9},"Changqing Ke",{"name":199,"orcid":9},"Bin Cheng",{"name":201,"orcid":9},"Zheng Duan",{"tldr":203,"method":204,"finding":205,"direction":64,"opportunity":206},"基于Transformer及变体融合多源卫星与地下水数据，将SMAP土壤湿度降尺度至1km日尺度。","Transformer、PatchTST、iTransformer，融合多源卫星","Transformer降尺度产品精度最优（ubRMSE=0.0372），保留原产品时空格局并响应降水","可探索Transformer降尺度产品在区域干旱监测、灌溉决策及作物估产中的耦合应用。","2026-09-23T23:30:19.132307Z",{"id":209,"title":210,"url":211,"summary":212,"summary_zh":213,"content":9,"source_name":214,"source_url":211,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":215,"score_detail":216,"sources":219,"tags":221,"search_phrases":224,"slug":227,"view_count":15,"doi":228,"paper":229,"created_at":247},3256,"Leakage-aware, calibrated, and explainable deep learning for robust almond disease classification","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112464","Leakage-aware, calibrated, and explainable deep learning for robust almond disease classification。Computers and Electronics in Agriculture","面向稳健杏仁病害分类的泄漏感知、校准且可解释的深度学习。《农业计算机与电子》","Computers and Electronics in Agriculture",72,{"impact":83,"substance":217,"depth":82,"authority":174,"freshness":175,"relevant":69,"comment":218},20,"核心期刊论文，方法上有防泄漏、校准与可解释性创新，但作物小众、属细分技术进展，未达每日精选门槛。",[220],{"name":214,"url":211},[136,89,21,222,223],"病害识别","巴旦木",[225,226],"巴旦木 病害 深度学习","农业人工智能 智慧农业 深度学习 病害识别","巴旦木病害深度学习-3256","10.1016\u002Fj.compag.2026.112464",{"doi":228,"openalex_id":230,"authors":231,"venue":214,"cited_by_count":15,"oa_url":211,"card":242,"direction":62,"ingested_from":65},"W7213988471",[232,235,238,240],{"name":233,"orcid":234},"Abebaw Degu Workneh","https:\u002F\u002Forcid.org\u002F0000-0001-7694-1577",{"name":236,"orcid":237},"Badr Elkari","https:\u002F\u002Forcid.org\u002F0000-0002-0893-783X",{"name":239,"orcid":9},"Meryam El Mouhtadi",{"name":241,"orcid":9},"Mohammad Furqan Ali",{"tldr":243,"method":244,"finding":245,"direction":62,"opportunity":246},"提出防泄漏、校准且可解释的深度学习框架，用于稳健的杏仁病害分类。","采用防数据泄漏的深度学习训练、概率校准与可解释性分析。","该框架能提升杏仁病害分类的稳健性、可信度与可解释性。","可探索防泄漏与校准机制在其他作物病害识别中的泛化及田间部署。","2026-09-23T23:30:01.628054Z",{"id":249,"title":250,"url":251,"summary":252,"summary_zh":253,"content":9,"source_name":254,"source_url":251,"published_at":255,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":256,"sources":258,"tags":260,"search_phrases":263,"slug":266,"view_count":15,"doi":267,"paper":268,"created_at":283},3188,"Identification of granitic pegmatites based on GF-5 hyperspectral data and LSTM-Transformer model","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.acags.2026.100406","Lithium is a key strategic mineral to support new energy and advanced technology, mainly enriched in granitic pegmatite. The integration of hyperspectral remote sensing and deep learning frameworks has emerged as a leading approach for quantitative identification of pegmatites. However, deep learning algorithms are often limited by the scarcity of labeled samples and the challenge of extracting features from high-dimensional spectral data. In this study, a data augmentation-driven LSTM-Transformer framework is proposed for pegmatite mapping using GF-5 hyperspectral remote sensing data. In view of limited samples, this study constructed an enhanced training data set by integrating field-measured and infrared scanning spectra, simulating real spectral variability and linear mixing to improve data diversity. The proposed LSTM-Transformer architecture combines LSTM’s advantages in local sequence feature modeling and Transformer’s capacity to capture long-range global relationships, thereby enhancing the detection of weakly diagnostic spectral features. The framework was demonstrated in the Jingerquan lithium deposit, Xinjiang, China. The high-probability areas delineated by the proposed model successfully captured several pegmatite dikes documented in previous geological studies. The prediction results were further validated through field sampling. Geochemical analyses of specimens collected from the predicted high-probability target areas show an average Li 2 O grade of 7.86 wt.%, which meets industrial requirements and confirms the reliability of the proposed method. Overall, these results provide the potential of the proposed framework for pegmatite mapping and subsequent mineral exploration under the conditions of limited labeled samples.","锂是支撑新能源与先进技术的关键战略矿产，主要富集于花岗伟晶岩中。高光谱遥感与深度学习框架的融合已成为伟晶岩定量识别的前沿方法。然而，深度学习算法常受限于标注样本稀缺以及高维光谱数据特征提取的难题。本研究提出了一种数据增强驱动的LSTM-Transformer框架，用于基于GF-5高光谱遥感数据的伟晶岩填图。针对样本有限的问题，本研究通过整合野外实测光谱与红外扫描光谱构建了增强训练数据集，模拟真实光谱变异与线性混合以提升数据多样性。所提出的LSTM-Transformer架构结合了LSTM在局部序列特征建模方面的优势与Transformer捕捉长程全局关系的能力，从而增强了对弱诊断光谱特征的检测。该框架在中国新疆镜儿泉锂矿床进行了验证。所提模型圈定的高概率区域成功捕获了以往地质研究中记录的若干伟晶岩脉。预测结果进一步通过野外采样得到验证。从预测高概率目标区采集的标本地球化学分析显示，Li₂O平均品位为7.86 wt.%，满足工业要求，证实了所提方法的可靠性。总体而言，这些结果展示了所提框架在标注样本有限条件下用于伟晶岩填图及后续矿产勘查的潜力。","Applied Computing and Geosciences","2026-09-19T00:00:00Z",{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":257},"该论文聚焦高光谱遥感与深度学习用于伟晶岩型锂矿识别，属地质矿产勘探领域，与三农、农业信息化、智慧农业主题无关，不建议进入每日精选。",[259],{"name":254,"url":251},[21,90,261,262],"高光谱遥感","锂矿勘探",[264,265],"GF-5 高光谱 伟晶岩","新疆 镜儿泉 锂矿","GF-5高光谱伟晶岩-3188","10.1016\u002Fj.acags.2026.100406",{"doi":267,"openalex_id":269,"authors":270,"venue":254,"cited_by_count":15,"oa_url":277,"card":278,"direction":64,"ingested_from":65},"W7213617576",[271,274],{"name":272,"orcid":273},"Zhong Li","https:\u002F\u002Forcid.org\u002F0000-0002-2312-3746",{"name":275,"orcid":276},"Ziye Wang","https:\u002F\u002Forcid.org\u002F0000-0001-6538-5798","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS259019742600090X\u002Fpdf",{"tldr":279,"method":280,"finding":281,"direction":64,"opportunity":282},"提出数据增强驱动的LSTM-Transformer框架，用GF-5高光谱数据识别花岗伟晶岩。","GF-5高光谱数据，融合实测与红外扫描光谱增强样本，构建LSTM-Transfo","在新疆金儿泉锂矿成功圈定伟晶岩脉，验证区Li2O平均品位7.86 wt.%。","可将该高光谱-深度学习框架迁移至农田土壤属性或作物养分反演，解决标签稀缺问题。","2026-09-22T23:30:25.608490Z",{"id":285,"title":286,"url":287,"summary":288,"summary_zh":289,"content":9,"source_name":290,"source_url":287,"published_at":291,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":292,"score_detail":293,"sources":295,"tags":297,"search_phrases":300,"slug":303,"view_count":15,"doi":304,"paper":305,"created_at":338},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":83,"substance":217,"depth":82,"authority":131,"freshness":175,"relevant":69,"comment":294},"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。",[296],{"name":290,"url":287},[136,89,21,298,299],"可可","作物表型",[301,302],"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":304,"openalex_id":306,"authors":307,"venue":290,"cited_by_count":15,"oa_url":287,"card":333,"direction":64,"ingested_from":65},"W7213901640",[308,310,313,315,317,319,322,325,328,330],{"name":309,"orcid":9},"Ezekiel Ahn",{"name":311,"orcid":312},"Eun-Sung Park","https:\u002F\u002Forcid.org\u002F0000-0001-6826-2865",{"name":314,"orcid":9},"Moon S. Kim",{"name":316,"orcid":9},"Hangi Kim",{"name":318,"orcid":9},"Lalit M. Kandpal",{"name":320,"orcid":321},"Sunchung Park","https:\u002F\u002Forcid.org\u002F0000-0002-7398-9476",{"name":323,"orcid":324},"Seunghyun Lim","https:\u002F\u002Forcid.org\u002F0000-0003-3023-4863",{"name":326,"orcid":327},"Lyndel W. Meinhardt","https:\u002F\u002Forcid.org\u002F0000-0001-8299-2629",{"name":329,"orcid":9},"Byoung-Kwan Cho",{"name":331,"orcid":332},"Insuck Baek","https:\u002F\u002Forcid.org\u002F0000-0003-1044-349X",{"tldr":334,"method":335,"finding":336,"direction":64,"opportunity":337},"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"]