[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-4079":3,"related-4079":123},{"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":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":43,"search_phrases":91,"slug":94,"view_count":95,"doi":96,"paper":97,"created_at":122},4079,"Pixel-level high-throughput estimation of crop photosynthetic phenotyping parameters using a multi-task deep learning framework","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.aiia.2026.08.014","Hyperspectral remote sensing holds promise for estimating crop physiological and biochemical traits, yet conventional approaches are often constrained by small sample sizes, limited accuracy in multi-trait retrieval, and poor robustness. To alleviate these challenges, we present an integrated framework that combines a pixel-wise dataset construction strategy with a multi-task deep learning architecture for estimating photosynthetic phenotyping parameters (PPPs)—a suite of traits encompassing gas exchange, pigment composition, and structural properties. Specifically, under controlled experimental conditions, we developed MTI-Net, a lightweight multi-task inversion network that simultaneously estimates multiple PPPs in rice and tomato with high precision ( R 2 > 0.90). Since physiological ground-truth measurements were obtained at the leaf level, the pixel-wise predictions should be interpreted as weakly supervised, model-derived spatial estimates or relative distributions rather than directly validated pixel-level physiological measurements. This methodological framework demonstrated stable performance across data partitions and random initializations, and maintained stable performance when applied to pixel-averaged spectra. A complementary differentiable band selection module (DRS-Net) further reduced spectral dimensionality to 40 informative bands while maintaining retrieval accuracy above 0.75, suggesting potential pathways for simplified sensor design. Importantly, the estimated PPPs revealed biologically interpretable patterns in tomato, including stage-dependent dynamics, irradiance-driven shifts, and canopy-layer differentiation—where middle leaves dominated early photosynthesis and upper leaves prevailed at later stages. Collectively, this study provides a methodological foundation for proximal hyperspectral phenotyping under specific environments. While further validation in diverse field conditions is required, the framework demonstrates potential for high-throughput multi-parameter estimation and provides a methodological basis for future cost-effective hyperspectral systems.","高光谱遥感在估算作物生理生化性状方面具有潜力，但传统方法常受限于样本量小、多性状反演精度有限以及鲁棒性差等问题。为缓解这些挑战，我们提出了一种集成框架，将像元级数据集构建策略与多任务深度学习架构相结合，用于估算光合表型参数（photosynthetic phenotyping parameters, PPPs）——一组涵盖气体交换、色素组成和结构特性的性状。具体而言，在受控实验条件下，我们开发了MTI-Net，一种轻量级多任务反演网络，可同时高精度估算水稻和番茄的多个PPPs（R² > 0.90）。由于生理真值测量是在叶片尺度获得的，像元级预测应被解释为弱监督的、模型导出的空间估计或相对分布，而非直接验证的像元级生理测量。该方法框架在数据划分和随机初始化下表现出稳定性能，并在应用于像元平均光谱时保持稳定性能。一个互补的可微分波段选择模块（DRS-Net）进一步将光谱维度降至40个信息波段，同时保持反演精度高于0.75，提示了简化传感器设计的潜在路径。重要的是，估算的PPPs在番茄中揭示了具有生物学可解释性的模式，包括阶段依赖性动态、辐照度驱动的变化以及冠层分层分化——其中中层叶片在早期光合作用中占主导，而上层叶片在后期阶段占优势。总体而言，本研究为特定环境下的近地高光谱表型分析提供了方法学基础。尽管需要在多样化田间条件下进一步验证，但该框架展示了高通量多参数估算的潜力，并为未来经济高效的高光谱系统提供了方法学依据。",null,"Artificial Intelligence in Agriculture","2026-10-01T00:00:00Z","论文",10,false,80,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,14,8,1,"方法新颖、结论可靠的高光谱作物表型多任务反演研究，对智慧农业遥感监测有参考价值。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业遥感","深度学习","高光谱","作物表型",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[38],{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},"I3","remote-sensing","遥感与农情监测","\u002Ftopic\u002Fdigital-agriculture\u002Fremote-sensing",{"status":44,"locked":14,"version":45,"method":46,"evidence":47},"classified","2026-10-22.2","rules",{"candidates":48,"agricultural_evidence":51},[49,56,64,72,79,86],{"code":32,"score":50,"supported":51,"title":52,"abstract":54,"tags":55},15,true,[53],"deep learning",[28,53],[28],{"code":39,"score":57,"supported":51,"title":58,"abstract":59,"tags":63},12,[],[60,61,62],"遥感","remote sensing","hyperspectral",[30,27,29],{"code":65,"score":66,"supported":51,"title":67,"abstract":68,"tags":71},"A1",6,[],[69,70],"水稻","rice",[],{"code":73,"score":66,"supported":51,"title":74,"abstract":75,"tags":78},"B1",[],[76,77],"番茄","tomato",[],{"code":80,"score":81,"supported":51,"title":82,"abstract":83,"tags":85},"I7",3,[],[84],"methodological framework",[],{"code":87,"score":21,"supported":14,"title":88,"abstract":89,"tags":90},"I1",[],[],[26],[92,93],"MTI-Net 高光谱 作物表型","水稻 番茄 光合表型 高光谱","MTI-Net高光谱作物表型-4079",5,"10.1016\u002Fj.aiia.2026.08.014",{"doi":96,"openalex_id":98,"authors":99,"venue":10,"cited_by_count":114,"oa_url":6,"card":115,"direction":119,"ingested_from":121},"W7215067733",[100,102,104,107,110,112],{"name":101,"orcid":9},"Mengqi Zhang",{"name":103,"orcid":9},"Jiaren Zhou",{"name":105,"orcid":106},"Qingfeng Song","https:\u002F\u002Forcid.org\u002F0000-0003-4314-6897",{"name":108,"orcid":109},"Man Zhang","https:\u002F\u002Forcid.org\u002F0009-0007-4481-0514",{"name":111,"orcid":9},"Xinguang Zhu",{"name":113,"orcid":9},"Minjuan Wang",0,{"tldr":116,"method":117,"finding":118,"direction":119,"opportunity":120},"提出像素级多任务深度学习框架，从高光谱数据高通量估算作物光合表型参数。","构建像素级数据集，设计轻量多任务网络MTI-Net及可微波段选择模块DRS-Ne","MTI-Net对水稻和番茄多个光合参数估算R²>0.90，DRS-Net降至40波段仍保持精度>0.","农业遥感与作物表型","可探索弱监督像素级预测的田间验证，以及低成本高光谱传感器与多任务模型的联合优化。","openalex","2026-10-03T23:30:24.577846Z",{"total":66,"page":21,"page_size":66,"items":124},[125,198,281,356,421,508],{"id":126,"title":127,"url":128,"summary":129,"summary_zh":130,"content":9,"source_name":131,"source_url":128,"published_at":132,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":133,"score_detail":134,"sources":139,"tags":141,"primary_topic":144,"secondary_topics":145,"classification":147,"search_phrases":170,"slug":173,"view_count":21,"doi":174,"paper":175,"created_at":197},5165,"Proximal Remote Sensing of Oil Palm Basal Stem Rot Using Deep Learning Vision Algorithms and Ground-based Images","https:\u002F\u002Fdoi.org\u002F10.22266\u002Fijies2026.1130.31","Basal Stem Rot (BSR), caused by Ganoderma boninense, threatens oil palm productivity in Southeast Asia.This study develops and benchmarks a low-cost remote sensing framework for BSR severity diagnosis from smartphone-captured RGB trunk images, evaluating two pathways for future field deployment.The first pairs YOLO detection with a Random Forest classifier: YOLO26n gave the best accuracy-to-cost trade-off among three nanovariants for localizing BSR symptoms (mushroom, rot, white-button), and the end-to-end YOLO26n+RF pipeline achieved 0.83 accuracy and macro-F1 of 0.79 across Healthy, Mild, Moderate, and Severe classes.Rot was consistently the hardest symptom to localize, owing to its diffuse, boundary-ambiguous appearance.The second pathway trained four image classifiers (EfficientNet-B0, MobileNetV2, Swin-Tiny, and ConvNeXt-V2-Tiny) directly on severity labels, with ConvNeXt-V2-Tiny performing best (0.83 accuracy, 0.83 macro-F1).Both pathways performed well at severity extremes but struggled with Moderate, reflecting visual overlap between stages.SHAP analysis identified rot-area ratio and fruiting-body count as top predictors.These offline results show accurate BSR diagnosis is potentially achievable from smartphone imagery, laying groundwork for future on-device deployment.","由灵芝（Ganoderma boninense）引起的基茎腐病（Basal Stem Rot, BSR）威胁着东南亚油棕的生产力。本研究开发并基准测试了一个低成本遥感框架，用于从智能手机拍摄的RGB树干图像中诊断基茎腐病的严重程度，并评估了未来田间部署的两条技术路径。第一条路径将YOLO检测与随机森林分类器相结合：在三种纳米变体中，YOLO26n在定位基茎腐病症状（蘑菇、腐烂、白钮）方面实现了最佳的精度-成本权衡，端到端的YOLO26n+RF流程在健康、轻度、中度和重度四个类别上达到了0.83的准确率和0.79的宏F1值。腐烂始终是最难定位的症状，原因在于其弥散且边界模糊的外观。第二条路径直接在严重程度标签上训练了四个图像分类器（EfficientNet-B0、MobileNetV2、Swin-Tiny和ConvNeXt-V2-Tiny），其中ConvNeXt-V2-Tiny表现最佳（准确率0.83，宏F1值0.83）。两条路径在严重程度极端类别上表现良好，但在中度类别上表现欠佳，反映出各阶段之间存在视觉重叠。SHAP分析确定腐烂面积比和子实体数量为最重要的预测因子。这些离线结果表明，利用智能手机图像有望实现准确的基茎腐病诊断，为未来设备端部署奠定了基础。","International journal of intelligent engineering and systems","2026-10-16T00:00:00Z",69,{"impact":57,"substance":135,"depth":136,"authority":57,"freshness":137,"relevant":21,"comment":138},20,16,9,"该研究利用智能手机图像与深度学习实现油棕基腐病严重度诊断，方法新颖、结论可靠，但属细分作物研究，产业影响有限，适合作为技术类精选。",[140],{"name":131,"url":128},[26,27,28,142,143],"油棕","病害诊断",{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},[146],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":44,"locked":14,"version":45,"method":46,"evidence":148},{"candidates":149,"agricultural_evidence":51},[150,154,159,166],{"code":39,"score":50,"supported":51,"title":151,"abstract":152,"tags":153},[61],[60,61],[27],{"code":32,"score":57,"supported":51,"title":155,"abstract":156,"tags":158},[53],[157],"YOLO",[28],{"code":160,"score":66,"supported":51,"title":161,"abstract":162,"tags":165},"B5",[],[163,164],"蘑菇","mushroom",[],{"code":87,"score":21,"supported":14,"title":167,"abstract":168,"tags":169},[],[],[26],[171,172],"油棕 基腐病 遥感","Ganoderma boninense 智能手机 检测","油棕基腐病遥感-5165","10.22266\u002Fijies2026.1130.31",{"doi":174,"openalex_id":176,"authors":177,"venue":131,"cited_by_count":114,"oa_url":128,"card":192,"direction":119,"ingested_from":121},"W7222626135",[178,181,184,186,189],{"name":179,"orcid":180},"Wan Siu Hong","https:\u002F\u002Forcid.org\u002F0000-0002-8254-4962",{"name":182,"orcid":183},"Jonathan Yong Chung Ee","https:\u002F\u002Forcid.org\u002F0000-0003-4079-8005",{"name":185,"orcid":9},"Jin Yuan Chan",{"name":187,"orcid":188},"Mohamad Izzuddin Anuar","https:\u002F\u002Forcid.org\u002F0000-0002-4929-5175",{"name":190,"orcid":191},"Mahmud Iwan Solihin","https:\u002F\u002Forcid.org\u002F0000-0002-5293-7466",{"tldr":193,"method":194,"finding":195,"direction":119,"opportunity":196},"用手机拍摄油棕树干图像，结合深度学习检测与分类，实现基腐病严重度诊断。","YOLO26n+随机森林与四种图像分类器，基于手机RGB图像，SHAP解释。","两条路径准确率均约0.83，中度病级最难区分，腐区比例与子实体数为关键预测因子。","可探索轻量化模型在手机端离线部署，并扩充多地区、多品种数据以提升中度病级辨识。","2026-10-17T23:30:28.307953Z",{"id":199,"title":200,"url":201,"summary":202,"summary_zh":203,"content":9,"source_name":204,"source_url":201,"published_at":205,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":206,"score_detail":207,"sources":212,"tags":214,"primary_topic":217,"secondary_topics":218,"classification":226,"search_phrases":252,"slug":255,"view_count":114,"doi":256,"paper":257,"created_at":280},4898,"An Interpretable Deep Learning Framework for Spatial Prediction of Soil Salinity Using Multi-Source Remote Sensing and Environmental Data: A Case Study of Cangzhou, Hebei Province, China","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fs26206465","Soil salinity poses a significant challenge to agricultural productivity and sustainable land management in coastal and semi-arid regions. Soil salinity in coastal agricultural regions is controlled by complex interactions among surface conditions, groundwater processes, soil properties, and terrain characteristics. Accurate spatial prediction requires integrating remote sensing observations with environmental information that represents these processes. This study presents a multi-source deep learning framework that integrates Sentinel-2-derived spectral indices with environmental covariates for regional soil salinity spatial prediction under the environmental conditions of the 2024 sampling period. SHapley Additive exPlanations (SHAP) is employed to quantify the contribution of environmental variables within the prediction framework and improve model interpretability. A convolutional neural network (CNN) incorporating a squeeze-and-excitation (SE) attention mechanism is developed as a nonlinear feature representation framework for integrating heterogeneous predictors, while the Crested Porcupine Optimizer (CPO) is used to optimize model hyperparameters. The proposed framework achieved the highest predictive performance among the benchmark models, with an R2 of 0.785, an RMSE of 0.928, and an MAE of 0.720. Integrating environmental covariates with spectral features further improves the prediction performance, increasing the R2 from 0.701 to 0.785. The results indicate the potential of integrating multi-source environmental information with interpretable deep learning for spatial prediction of near-surface soil salinity under similar coastal agricultural conditions.","土壤盐渍化对沿海和半干旱地区的农业生产力和可持续土地管理构成了重大挑战。沿海农业区域的土壤盐渍化受地表条件、地下水过程、土壤属性和地形特征之间复杂相互作用的控制。准确的空间预测需要将遥感观测与表征这些过程的环境信息相结合。本研究提出了一种多源深度学习框架，该框架将Sentinel-2衍生的光谱指数与环境协变量相结合，用于2024年采样期环境条件下的区域土壤盐渍化空间预测。采用SHapley加法解释（SHAP）方法量化预测框架中环境变量的贡献并提高模型可解释性。开发了一种融合压缩-激励（SE）注意力机制的卷积神经网络（CNN）作为非线性特征表示框架，用于整合异质性预测因子，同时使用冠豪猪优化器（CPO）优化模型超参数。所提出的框架在基准模型中取得了最高的预测性能，R²为0.785，RMSE为0.928，MAE为0.720。将环境协变量与光谱特征相结合进一步提高了预测性能，R²从0.701提升至0.785。结果表明，在类似的沿海农业条件下，将多源环境信息与可解释深度学习相结合用于近地表土壤盐渍化空间预测具有潜力。","Sensors","2026-10-13T00:00:00Z",75,{"impact":136,"substance":208,"depth":209,"authority":210,"freshness":20,"relevant":21,"comment":211},21,17,13,"方法新颖、指标可靠，将多源遥感与可解释深度学习结合用于盐渍化空间预测，对沿海农业区土壤治理有参考价值。",[213],{"name":204,"url":201},[26,27,215,28,216],"可解释AI","土壤盐渍化",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[219,220],{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},{"code":221,"slug":222,"name":223,"path":224,"parent_code":225},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",{"status":44,"locked":14,"version":45,"method":46,"evidence":227},{"candidates":228,"agricultural_evidence":51},[229,233,237,242,248],{"code":32,"score":136,"supported":51,"title":230,"abstract":231,"tags":232},[53],[28,53],[215,28],{"code":39,"score":50,"supported":51,"title":234,"abstract":235,"tags":236},[61],[60,61],[27],{"code":221,"score":19,"supported":51,"title":238,"abstract":239,"tags":241},[222],[240,222],"土壤",[],{"code":243,"score":81,"supported":51,"title":244,"abstract":245,"tags":247},"H7",[],[246],"干旱",[],{"code":87,"score":21,"supported":14,"title":249,"abstract":250,"tags":251},[],[],[26],[253,254],"沧州 土壤盐渍化 遥感","Sentinel-2 土壤盐分 预测","沧州土壤盐渍化遥感-4898","10.3390\u002Fs26206465",{"doi":256,"openalex_id":258,"authors":259,"venue":204,"cited_by_count":114,"oa_url":201,"card":275,"direction":119,"ingested_from":121},"W7221715654",[260,263,266,268,271,273],{"name":261,"orcid":262},"Mingyuan Ye","https:\u002F\u002Forcid.org\u002F0009-0001-6434-5505",{"name":264,"orcid":265},"Peng Chen","https:\u002F\u002Forcid.org\u002F0000-0002-2549-6765",{"name":267,"orcid":9},"Junchi Ma",{"name":269,"orcid":270},"Bingxia Liu","https:\u002F\u002Forcid.org\u002F0000-0001-7958-7112",{"name":272,"orcid":9},"Jiahao Qi",{"name":274,"orcid":9},"Yuan Jing",{"tldr":276,"method":277,"finding":278,"direction":119,"opportunity":279},"构建融合多源遥感与环境变量的可解释深度学习框架，预测沧州土壤盐分空间分布。","Sentinel-2光谱指数+环境协变量，CNN-SE注意力，CPO优化超参数，","环境协变量与光谱特征结合将R²从0.701提升至0.785，模型性能最优。","可迁移至其他滨海盐渍区，并探索时序动态预测与多深度土壤盐分协同反演。","2026-10-14T23:30:27.313838Z",{"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":292,"tags":294,"primary_topic":298,"secondary_topics":304,"classification":307,"search_phrases":333,"slug":336,"view_count":95,"doi":337,"paper":338,"created_at":355},4433,"PROSAIL-DNN-TL: A Hyperspectral Inversion Framework for Winter Wheat Leaf Chlorophyll Content Based on Fine-Tuned Transfer Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagronomy16191971","Accurate and efficient estimation of leaf chlorophyll content (LCC) is essential for crop physiological monitoring and precision agriculture. Current remote sensing approaches for LCC estimation mainly rely on machine learning and deep learning methods, which exhibit strong nonlinear modelling capabilities but require sufficient and representative samples. Radiative transfer models (RTMs) can simulate canopy spectral response based on physical mechanisms and generate large-scale datasets, thereby overcoming the limitations of purely data-driven approaches. However, their applications under diverse environmental conditions remain challenging due to uncertainties caused by model parameterization and discrepancies between simulated and observed data. This study proposed a hybrid framework, PROSAIL-DNN-TL, integrating the PROSAIL RTM with full-network fine-tuned transfer learning for winter wheat LCC estimation. First, 85,680 simulated samples generated by PROSAIL were used to pre-train a deep neural network (DNN), and field observations (n = 270) were subsequently used for full-network fine-tuning. A conventional DNN model trained only with field data and machine learning models based on different feature selection strategies were established for comparison. The results showed that: (1) UVE-selected bands were distributed across visible, red-edge, and near-infrared regions; CARS retained more bands in the visible region, whereas SPA mainly selected near-infrared bands; (2) RR and SVR achieved higher accuracy than KNN, with the full-spectrum and CARS feature sets showing superior performance and SPA showing the lowest accuracy; (3) PROSAIL-DNN-TL achieved reliable LCC estimation performance, with an average test R2 of 0.79 across 500 repeated experiments, and showed improved performance compared with conventional machine learning models; (4) PROSAIL-DNN-TL achieved reliable predictive performance when evaluated using datasets collected during the 2024 and 2025 growing seasons, with R2 values of 0.75 and 0.92, respectively. Overall, the proposed hybrid framework integrates physical knowledge from RTMs with data-driven modelling through transfer learning, providing a potential approach for crop chlorophyll monitoring. However, the SPAD-to-LCC conversion used in this study was not specifically calibrated for the instrument and crop, and therefore, the absolute LCC values should be interpreted with caution.","准确、高效地估算叶片叶绿素含量（leaf chlorophyll content，LCC）对作物生理监测和精准农业至关重要。当前遥感估算LCC的方法主要依赖机器学习和深度学习方法，这类方法具有较强的非线性建模能力，但需要充足且具有代表性的样本。辐射传输模型（radiative transfer models，RTMs）能够基于物理机制模拟冠层光谱响应并生成大规模数据集，从而克服纯数据驱动方法的局限性。然而，由于模型参数化带来的不确定性以及模拟数据与观测数据之间的差异，其在不同环境条件下的应用仍面临挑战。本研究提出了一种混合框架PROSAIL-DNN-TL，将PROSAIL辐射传输模型与全网络微调迁移学习相结合，用于冬小麦LCC估算。首先，利用PROSAIL生成的85 680个模拟样本预训练深度神经网络（deep neural network，DNN），随后利用田间观测数据（n = 270）进行全网络微调。同时建立了仅使用田间数据训练的传统DNN模型以及基于不同特征选择策略的机器学习模型作为对比。结果表明：（1）UVE选择的波段分布在可见光、红边和近红外区域；CARS在可见光区域保留了更多波段，而SPA主要选择近红外波段；（2）RR和SVR的精度高于KNN，全光谱和CARS特征集表现较优，SPA精度最低；（3）PROSAIL-DNN-TL取得了可靠的LCC估算性能，在500次重复实验中平均测试R²为0.79，且相比传统机器学习模型表现出更优的性能；（4）利用2024年和2025年生长季采集的数据集进行评估时，PROSAIL-DNN-TL均表现出可靠的预测性能，R²分别为0.75和0.92。总体而言，所提出的混合框架通过迁移学习将辐射传输模型的物理知识与数据驱动建模相结合，为作物叶绿素监测提供了一种潜在途径。然而，本研究中使用的SPAD到LCC的转换并未针对仪器和作物进行专门校准，因此绝对LCC值应谨慎解读。","Agronomy","2026-10-07T00:00:00Z",81,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":137,"relevant":21,"comment":291},"将PROSAIL辐射传输模型与深度神经网络迁移学习结合，用8万余模拟样本预训练、270个田间样本微调，冬小麦叶绿素反演精度可靠，方法新颖且数据规模扎实，对作物生理监测与精准农业有参考价值。",[293],{"name":287,"url":284},[26,27,295,296,297],"冬小麦","迁移学习","叶绿素监测",{"code":299,"slug":300,"name":301,"path":302,"parent_code":303},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[305,306],{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":44,"locked":14,"version":45,"method":46,"evidence":308},{"candidates":309,"agricultural_evidence":51},[310,316,320,326],{"code":299,"score":311,"supported":51,"title":312,"abstract":314,"tags":315},26,[313,300],"winter wheat",[301,313,300],[295],{"code":39,"score":50,"supported":51,"title":317,"abstract":318,"tags":319},[62],[60,61],[27],{"code":32,"score":13,"supported":51,"title":321,"abstract":322,"tags":325},[],[28,323,324,53],"机器学习","machine learning",[296],{"code":87,"score":327,"supported":51,"title":328,"abstract":329,"tags":332},7,[],[330,331],"精准农业","precision agriculture",[26],[334,335],"PROSAIL 冬小麦 叶绿素","高光谱 叶绿素 反演","PROSAIL冬小麦叶绿素-4433","10.3390\u002Fagronomy16191971",{"doi":337,"openalex_id":339,"authors":340,"venue":287,"cited_by_count":114,"oa_url":284,"card":350,"direction":119,"ingested_from":121},"W7220844556",[341,343,345,348],{"name":342,"orcid":9},"Xueqing Zhu",{"name":344,"orcid":9},"Ying Nian",{"name":346,"orcid":347},"Jikai Liu","https:\u002F\u002Forcid.org\u002F0000-0003-2365-5221",{"name":349,"orcid":9},"Xinwei Li",{"tldr":351,"method":352,"finding":353,"direction":119,"opportunity":354},"提出PROSAIL-DNN-TL框架，用PROSAIL模拟数据预训练DNN并全网络微调，反演冬小麦叶","PROSAIL辐射传输模型生成85680样本预训练DNN，270个田间样本全网络","500次重复实验平均测试R²为0.79，2024和2025年生长季验证R²分别为0.75和0.92，","SPAD到LCC转换未针对仪器和作物标定，可研究跨传感器、跨作物、跨生长季的迁移学习泛化与绝对叶绿素标定。","2026-10-08T23:30:29.383957Z",{"id":357,"title":358,"url":359,"summary":360,"summary_zh":361,"content":9,"source_name":362,"source_url":359,"published_at":363,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":364,"score_detail":365,"sources":367,"tags":369,"primary_topic":373,"secondary_topics":374,"classification":381,"search_phrases":402,"slug":405,"view_count":81,"doi":406,"paper":407,"created_at":420},3839,"A remote sensing image processing method based on edge computing and deep learning","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs43926-026-00508-2","The rapid advancement of Artificial Intelligence (AI) and edge computing has increased the demand for accurate and low-latency remote sensing image classification; however, conventional cloud-centric approaches face challenges related to processing latency, communication overhead, and efficient analysis of diverse land-cover scenes. This research proposes a remote sensing image processing method based on edge computing and deep learning (DL), integrating an Adaptive Chameleon Swarm-tuned Efficient Three-Dimensional Convolutional Neural Network (ACS-E3DCNN). The Remote Sensing Image Dataset obtained from Kaggle contains 5,631 RGB images categorized into five land-cover classes, namely agriculture, barren land, urban, vegetation, and water. During preprocessing, image normalization, adaptive histogram equalization, and non-local means filtering are applied to reduce noise, enhance contrast, and improve image quality. Subsequently, Principal Component Analysis (PCA) is employed for dimensionality refinement. The Efficient Three-Dimensional Convolutional Neural Network captures discriminative image representations with reduced parameter complexity, while the Adaptive Chameleon Swarm algorithm optimizes critical hyperparameters, including learning rate, kernel depth, and filter dimensions, to improve convergence and classification performance. Experimental results implemented using Python 3.10.1 demonstrate that the proposed model outperforms existing approaches, achieving an accuracy of 98.10%, F1-score of 96.85%, PSNR of 38.05, SSIM of 0.97, and latency of 85 ms under the 80:20 data split, with a statistically significant paired t-test result ( p \u003C 0.05). The proposed model provides an intelligent, reliable, and efficient approach for remote sensing image classification, supporting accurate land-cover mapping, environmental monitoring, and low-latency edge-enabled image processing.","人工智能（AI）与边缘计算的快速发展，提升了对高精度、低延迟遥感图像分类的需求；然而，传统的以云为中心的方法在处理延迟、通信开销以及高效分析多样化土地覆盖场景方面面临挑战。本研究提出了一种基于边缘计算与深度学习（DL）的遥感图像处理方法，集成了自适应变色龙群优化高效三维卷积神经网络（ACS-E3DCNN）。从Kaggle获取的遥感图像数据集包含5，631幅RGB图像，分为农业、裸地、城市、植被和水体五个土地覆盖类别。在预处理阶段，采用图像归一化、自适应直方图均衡化和非局部均值滤波来降低噪声、增强对比度并提升图像质量。随后，采用主成分分析（PCA）进行降维优化。高效三维卷积神经网络以较低的参数复杂度捕获具有判别性的图像表征，而自适应变色龙群算法则优化关键超参数，包括学习率、卷积核深度和滤波器尺寸，以改善收敛性和分类性能。使用Python 3.10.1实现的实验结果表明，所提模型优于现有方法，在80：20数据划分下达到了98.10%的准确率、96.85%的F1分数、38.05的PSNR、0.97的SSIM以及85 ms的延迟，配对t检验结果具有统计学显著性（p \u003C 0.05）。所提模型为遥感图像分类提供了一种智能、可靠且高效的方法，支持精确的土地覆盖制图、环境监测以及低延迟的边缘图像处理。","Discover Internet of Things","2026-09-29T00:00:00Z",67,{"impact":57,"substance":17,"depth":136,"authority":57,"freshness":137,"relevant":21,"comment":366},"该论文提出边缘计算与深度学习结合的遥感图像分类方法，在农业土地覆盖识别上精度高、延迟低，对智慧农业遥感应用有参考价值，但属方法类研究，产业影响有限。",[368],{"name":362,"url":359},[26,370,27,371,372],"农业人工智能","边缘计算","土地覆盖分类",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[375,376],{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},{"code":377,"slug":378,"name":379,"path":380,"parent_code":36},"I5","agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"status":44,"locked":14,"version":45,"method":46,"evidence":382},{"candidates":383,"agricultural_evidence":51},[384,389,393,398],{"code":32,"score":17,"supported":51,"title":385,"abstract":386,"tags":388},[53],[28,53,387],"artificial intelligence",[370],{"code":39,"score":50,"supported":51,"title":390,"abstract":391,"tags":392},[61],[60,61],[27],{"code":377,"score":50,"supported":51,"title":394,"abstract":396,"tags":397},[395],"edge computing",[371,395],[371],{"code":87,"score":21,"supported":14,"title":399,"abstract":400,"tags":401},[],[],[26],[403,404],"遥感图像 土地覆盖 分类","边缘计算 深度学习 遥感","遥感图像土地覆盖分类-3839","10.1007\u002Fs43926-026-00508-2",{"doi":406,"openalex_id":408,"authors":409,"venue":362,"cited_by_count":114,"oa_url":359,"card":415,"direction":119,"ingested_from":121},"W7214769118",[410,412],{"name":411,"orcid":9},"Yuli Song",{"name":413,"orcid":414},"Hong Zhang","https:\u002F\u002Forcid.org\u002F0000-0003-3176-3994",{"tldr":416,"method":417,"finding":418,"direction":119,"opportunity":419},"提出边缘计算与深度学习结合的遥感图像分类方法，用改进3D CNN实现低延迟高精度土地覆盖识别。","ACS-E3DCNN模型，结合自适应变色龙群优化超参数，用Kaggle 5631","准确率98.10%、F1为96.85%、延迟85ms，优于现有方法，支持低延迟边缘处理。","可探索边缘端轻量化模型在农田实时监测中的部署，结合多光谱与时序数据提升作物分类泛化性。","2026-09-30T23:30:27.868009Z",{"id":422,"title":423,"url":424,"summary":425,"summary_zh":426,"content":9,"source_name":427,"source_url":424,"published_at":428,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":429,"score_detail":430,"sources":432,"tags":434,"primary_topic":436,"secondary_topics":438,"classification":441,"search_phrases":461,"slug":464,"view_count":20,"doi":465,"paper":466,"created_at":507},3278,"Research on the optimal modeling path for inversion of Pb content in rice leaves based on hyperspectral data of ground objects and machine learning and cross-scale remote sensing monitoring","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs10661-026-15959-x","Research on the optimal modeling path for inversion of Pb content in rice leaves based on hyperspectral data of ground objects and machine learning and cross-scale remote sensing monitoring。Environmental Monitoring and Assessment","基于地物高光谱数据与机器学习的稻叶铅含量反演最优建模路径及跨尺度遥感监测研究。环境监测与评估","Environmental Monitoring and Assessment","2026-09-22T00:00:00Z",64,{"impact":20,"substance":17,"depth":136,"authority":210,"freshness":137,"relevant":21,"comment":431},"基于地物高光谱与机器学习的水稻叶片铅含量反演建模研究，方法有创新但属细分领域学术进展，公共影响有限。",[433],{"name":427,"url":424},[26,27,69,323,435],"高光谱遥感",{"code":65,"slug":70,"name":69,"path":437,"parent_code":303},"\u002Ftopic\u002Ffield-crops\u002Frice",[439,440],{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":44,"locked":14,"version":45,"method":46,"evidence":442},{"candidates":443,"agricultural_evidence":51},[444,449,453,457],{"code":39,"score":445,"supported":51,"title":446,"abstract":447,"tags":448},27,[61,62],[60,61,62],[27,435],{"code":32,"score":50,"supported":51,"title":450,"abstract":451,"tags":452},[324],[323,324],[323],{"code":65,"score":57,"supported":51,"title":454,"abstract":455,"tags":456},[70],[70],[69],{"code":87,"score":21,"supported":14,"title":458,"abstract":459,"tags":460},[],[],[26],[462,463],"水稻叶片 铅含量 高光谱 反演","稻米 重金属 遥感 监测","水稻叶片铅含量高光谱反演-3278","10.1007\u002Fs10661-026-15959-x",{"doi":465,"openalex_id":467,"authors":468,"venue":427,"cited_by_count":114,"oa_url":9,"card":502,"direction":119,"ingested_from":121},"W7214027889",[469,472,475,478,481,484,486,488,491,494,496,498,500],{"name":470,"orcid":471},"Zhenlong Zhang","https:\u002F\u002Forcid.org\u002F0009-0008-2354-9123",{"name":473,"orcid":474},"Zhe Wang","https:\u002F\u002Forcid.org\u002F0000-0003-1266-7251",{"name":476,"orcid":477},"Chengxia Wang","https:\u002F\u002Forcid.org\u002F0009-0001-0820-2462",{"name":479,"orcid":480},"Wenxue Lin","https:\u002F\u002Forcid.org\u002F0000-0002-8245-9063",{"name":482,"orcid":483},"Jingyan Zhang","https:\u002F\u002Forcid.org\u002F0009-0004-4567-5316",{"name":485,"orcid":9},"Ying Luo",{"name":487,"orcid":9},"Jiaqian Zhang",{"name":489,"orcid":490},"Kai Ye","https:\u002F\u002Forcid.org\u002F0000-0002-2851-6741",{"name":492,"orcid":493},"Yiming Chen","https:\u002F\u002Forcid.org\u002F0000-0002-8121-3109",{"name":495,"orcid":9},"Chaoliang Peng",{"name":497,"orcid":9},"Duan Tian",{"name":499,"orcid":9},"Weihao Wang",{"name":501,"orcid":9},"Jiaxin Liu",{"tldr":503,"method":504,"finding":505,"direction":119,"opportunity":506},"研究基于地面高光谱与机器学习反演水稻叶片铅含量，并探索跨尺度遥感监测的最优建模路径。","地面高光谱数据结合机器学习建模，开展跨尺度遥感监测。","明确了水稻叶片铅含量反演的最优建模路径，实现跨尺度遥感监测。","可探索多尺度遥感数据融合与迁移学习，提升重金属胁迫反演的普适性与精度。","2026-09-23T23:30:19.291361Z",{"id":509,"title":510,"url":511,"summary":512,"summary_zh":513,"content":9,"source_name":514,"source_url":511,"published_at":515,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":516,"sources":518,"tags":520,"primary_topic":521,"secondary_topics":522,"classification":525,"search_phrases":546,"slug":549,"view_count":550,"doi":551,"paper":552,"created_at":568},3164,"Spatiotemporal Deep Learning for Rice Plant Height Estimation from Multi-Temporal UAV RGB Imagery","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagriculture16182034","Accurate plant height estimation is important for monitoring crop growth and supporting precision agricultural management. Manual measurements are labor-intensive, while LiDAR-based methods are expensive and require complex processing. UAV photogrammetry provides a lower-cost alternative but remains challenging in flooded rice paddies because of canopy deformation and difficulties in terrain extraction. This study proposes Rice-STNet, a spatiotemporal deep learning framework for end-to-end rice plant height estimation using multi-temporal UAV RGB imagery. Rice-STNet integrates a convolutional neural network for spatial feature extraction, Time2Vec for temporal encoding, and a gated recurrent unit network for modeling temporal dependencies across observation dates. The framework was evaluated using field data collected from rice paddies over two growing seasons. Rice-STNet achieved an R2 of 0.97, a root mean squared error of 1.97 cm, and a mean absolute error of 1.14 cm. It outperformed random forest, support vector regression, a CNN-only baseline, and a UAV photogrammetry-based point-cloud approach. In addition, the framework generated high-resolution plant height maps for field-scale analysis of spatial growth variability. These results underscore the importance of jointly modeling spatial and temporal characteristics for continuously evolving crop traits. The proposed framework offers an accurate, scalable, and non-destructive solution for large-scale crop phenotyping and precision agriculture.","准确的株高估算对于监测作物生长和支持精准农业管理具有重要意义。人工测量劳动强度大，而基于激光雷达（LiDAR）的方法成本高昂且需要复杂的处理。无人机摄影测量提供了一种成本较低的替代方案，但在淹水稻田中仍面临挑战，原因在于冠层变形和地形提取困难。本研究提出了Rice-STNet，一种时空深度学习框架，用于利用多时相无人机RGB影像进行端到端水稻株高估算。Rice-STNet集成了用于空间特征提取的卷积神经网络、用于时间编码的Time2Vec，以及用于建模观测日期之间时间依赖关系的门控循环单元网络。该框架利用两个生长季从稻田采集的田间数据进行了评估。Rice-STNet取得了R²为0.97、均方根误差为1.97 cm、平均绝对误差为1.14 cm的结果。其性能优于随机森林、支持向量回归、仅使用CNN的基线方法以及基于无人机摄影测量的点云方法。此外，该框架生成了高分辨率株高图，用于田块尺度空间生长变异性分析。这些结果凸显了联合建模空间与时间特征对于持续变化的作物性状的重要性。所提出的框架为大规模作物表型分析和精准农业提供了一种准确、可扩展且非破坏性的解决方案。","Agriculture","2026-09-21T00:00:00Z",{"impact":17,"substance":18,"depth":17,"authority":210,"freshness":137,"relevant":21,"comment":517},"提出时空深度学习框架Rice-STNet，用多时相无人机RGB影像实现水稻株高高精度估算，方法新颖、数据跨两个生长季，对作物表型与精准农业有实用价值。",[519],{"name":514,"url":511},[26,370,69,60,30],{"code":65,"slug":70,"name":69,"path":437,"parent_code":303},[523,524],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"code":39,"slug":40,"name":41,"path":42,"parent_code":36},{"status":44,"locked":14,"version":45,"method":46,"evidence":526},{"candidates":527,"agricultural_evidence":51},[528,533,537,542],{"code":65,"score":17,"supported":51,"title":529,"abstract":530,"tags":532},[70],[69,531,70],"稻田",[69],{"code":32,"score":50,"supported":51,"title":534,"abstract":535,"tags":536},[53],[28,53],[370],{"code":39,"score":20,"supported":51,"title":538,"abstract":539,"tags":541},[],[30,540],"crop phenotyping",[30,60],{"code":87,"score":327,"supported":51,"title":543,"abstract":544,"tags":545},[],[330,331],[26],[547,548],"无人机 RGB 水稻株高","Rice-STNet 水稻表型","无人机RGB水稻株高-3164",2,"10.3390\u002Fagriculture16182034",{"doi":551,"openalex_id":553,"authors":554,"venue":514,"cited_by_count":114,"oa_url":511,"card":563,"direction":119,"ingested_from":121},"W7213887432",[555,558,560],{"name":556,"orcid":557},"Weiguo Wang","https:\u002F\u002Forcid.org\u002F0009-0003-4028-9363",{"name":559,"orcid":9},"Noboru Noguchi",{"name":561,"orcid":562},"Liangliang Yang","https:\u002F\u002Forcid.org\u002F0000-0002-5055-3987",{"tldr":564,"method":565,"finding":566,"direction":119,"opportunity":567},"提出Rice-STNet时空深度学习框架，用多时相无人机RGB影像估算水稻株高。","CNN提取空间特征，Time2Vec编码时间，GRU建模时序依赖，两季稻田数据验","R²达0.97、RMSE 1.97cm，优于随机森林、SVR、纯CNN及点云方法。","可迁移至其他作物与多源遥感融合，探索轻量化模型及实时田间部署。","2026-09-22T23:30:18.545958Z"]