[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3485":3,"related-3485":69},{"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,"search_phrases":32,"slug":35,"view_count":36,"doi":37,"paper":38,"created_at":68},3485,"Detection of Red Crown Rot of Soybean in Illinois Fields Using High-Resolution Satellite Imagery and Machine Learning","https:\u002F\u002Fdoi.org\u002F10.64898\u002F2026.09.22.753594","Red crown rot (RCR), caused by Calonectria ilicicola, is an emerging soybean disease in the U.S. Midwest for which scalable approaches to characterize within-field disease distribution are lacking. This study evaluated high-resolution PlanetScope satellite imagery for mapping RCR-affected soybean canopies across 15 commercial fields in Illinois surveyed during the 2024 and 2025 growing seasons. A total of 2,921 georeferenced canopy plots were classified as asymptomatic or RCR-affected and paired with six multispectral bands and seven vegetation indices. Spectral differences between classes were evaluated using linear mixed-effects models, and seven machine-learning classifiers representing linear, tree-based, neural-network, kernel, and probabilistic approaches were compared using spatially independent leave-one-field-out cross-validation. RCR-affected canopies exhibited increased reflectance in the visible and red-edge regions, reduced near-infrared reflectance, and lower vegetation-index values relative to asymptomatic canopies. All classifiers showed strong discrimination, with ROC-AUC values ranging from 0.963 to 0.982. Regularized logistic regression achieved the highest overall performance, with an accuracy of 0.945, balanced accuracy of 0.945, F1-score of 0.948, and ROC-AUC of 0.982 at the optimized decision threshold. Permutation analysis identified EVI, NDVI, and red reflectance as the most influential predictors across representative model architectures. Satellite-derived probability and classification maps generally corresponded with symptomatic canopy patterns observed in high-resolution UAV imagery, although mixed pixels reduced precision near disease-patch boundaries. These results demonstrate the potential of high-resolution satellite imagery for within-field mapping of RCR-associated canopy symptoms across independent commercial soybean fields.","由冬青丽赤壳菌（Calonectria ilicicola）引起的红冠腐病（RCR）是美国中西部一种新发大豆病害，目前尚缺乏可规模化表征田块内病害分布的方法。本研究评估了高分辨率PlanetScope卫星影像在2024年和2025年生长季对伊利诺伊州15块商业田块中受RCR影响的大豆冠层的制图能力。共有2,921个地理参考冠层样区被分类为无症状或受RCR影响，并与6个多光谱波段和7个植被指数配对。采用线性混合效应模型评估类别间的光谱差异，并通过空间独立的留一田块交叉验证比较了7种机器学习分类器，涵盖线性、基于树、神经网络、核函数和概率方法。与无症状冠层相比，受RCR影响的冠层在可见光和红边区域反射率增加，近红外反射率降低，植被指数值较低。所有分类器均表现出较强的判别能力，ROC-AUC值范围为0.963至0.982。正则化逻辑回归在优化决策阈值下取得了最高的整体性能，准确率为0.945，平衡准确率为0.945，F1分数为0.948，ROC-AUC为0.982。置换分析表明，在代表性模型架构中，EVI、NDVI和红光反射率是最具影响力的预测变量。卫星衍生的概率图和分类图总体上与高分辨率无人机影像中观察到的症状冠层模式一致，尽管混合像元降低了病害斑块边界附近的精度。这些结果证明了高分辨率卫星影像在独立商业大豆田块中对RCR相关冠层症状进行田块内制图的潜力。",null,"bioRxiv (Cold Spring Harbor Laboratory)","2026-09-23T00:00:00Z","论文",10,false,76,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,22,18,12,8,1,"基于高分辨率卫星影像与机器学习实现大豆红冠腐病田间制图，方法扎实、数据规模可观，对作物病害遥感监测有参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","机器学习","大豆","遥感监测","病害识别",[33,34],"伊利诺伊 大豆 红冠腐病 卫星遥感","PlanetScope 大豆 病害 机器学习","伊利诺伊大豆红冠腐病卫星遥感-3485",0,"10.64898\u002F2026.09.22.753594",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":60,"card":61,"direction":65,"ingested_from":67},"W7214126630",[41,44,47,49,51,54,57],{"name":42,"orcid":43},"Bruno Daniel Pugliese","https:\u002F\u002Forcid.org\u002F0009-0000-4795-4248",{"name":45,"orcid":46},"Juan Andrés Paredes","https:\u002F\u002Forcid.org\u002F0000-0002-3967-0197",{"name":48,"orcid":9},"Andres Fabian Ruiz",{"name":50,"orcid":9},"Norman Denis Bowman",{"name":52,"orcid":53},"Elhan S. Ersoz","https:\u002F\u002Forcid.org\u002F0000-0001-6930-6946",{"name":55,"orcid":56},"Nicolas Federico Martin","https:\u002F\u002Forcid.org\u002F0000-0002-1587-665X",{"name":58,"orcid":59},"Boris X. Camiletti","https:\u002F\u002Forcid.org\u002F0000-0002-1492-7988","https:\u002F\u002Fwww.biorxiv.org\u002Fcontent\u002Fbiorxiv\u002Fearly\u002F2026\u002F09\u002F23\u002F2026.09.22.753594.full.pdf",{"tldr":62,"method":63,"finding":64,"direction":65,"opportunity":66},"利用高分辨率卫星影像和机器学习在大豆田中检测红冠腐病。","PlanetScope多光谱影像、植被指数与7种机器学习分类器，留一田块交叉验证","各分类器ROC-AUC达0.963-0.982，正则化逻辑回归最优，EVI、NDVI和红波段最重要。","农业遥感与作物表型","可探索多时相卫星影像与无人机融合，提升病害斑块边界混合像元识别精度。","openalex","2026-09-25T23:30:22.839600Z",{"total":70,"page":22,"page_size":70,"items":71},6,[72,118,147,193,244,282],{"id":73,"title":74,"url":75,"summary":76,"summary_zh":77,"content":9,"source_name":78,"source_url":75,"published_at":79,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":80,"score_detail":81,"sources":87,"tags":89,"search_phrases":92,"slug":95,"view_count":36,"doi":96,"paper":97,"created_at":117},3499,"Tillage Practice Discrimination Using High-Resolution PlanetScope and Sentinel-2 Imagery","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs12524-026-02580-1","Abstract Accurate and timely assessment of soil tillage practices is crucial for monitoring sustainability in agriculture. To achieve more site-specific discrimination, it is necessary to understand the spectral and temporal properties of tillage practices across seasons. This study presents developments in tillage-practice discrimination by comparing two high-resolution remote sensing datasets, PlanetScope and Sentinel-2, to characterise and discriminate fields under intensive tillage (IT) and conservation tillage (CT) in the winter and spring seasons. A field experiment was conducted at an experimental site in the United Kingdom, collecting data on tillage practices in 2022–23 and 2023–24. We analysed the spectral and temporal characteristics of two tillage types and subsequently classified them using the random forest (RF) algorithm. Results showed reflectance differences between the two tillage treatments during the early period in both seasons. We also revealed that green, red-edge, and near-infrared wavelengths were relevant for the classification. PlanetScope showed greater potential for classifying tillage (OA = 70–80%), whereas Sentinel-2 exhibited lower performance (OA = 53–73%). Models from the winter achieved higher accuracy scores than those from the spring period, suggesting a seasonal variation in tillage discrimination. The findings highlight the utility of high-resolution satellite-based data, combined with machine learning, for mapping tillage practices and advancing precision agriculture.","准确的土壤耕作方式评估对于监测农业可持续性至关重要。为了实现更具针对性的区分，有必要了解不同季节耕作方式的光谱和时间特征。本研究通过比较两套高分辨率遥感数据集PlanetScope和Sentinel-2，在冬季和春季对集约耕作（IT）和保护性耕作（CT）田块进行表征与区分，从而推动耕作方式判别研究的发展。在英国一个试验站点开展了田间试验，收集了2022—23年和2023—24年的耕作方式数据。我们分析了两种耕作类型的光谱和时间特征，随后使用随机森林（RF）算法对其进行分类。结果表明，在兩個季节的早期阶段，两种耕作处理之间存在反射率差异。我们还发现，绿光、红边和近红外波段与分类相关。PlanetScope在耕作分类方面表现出更大的潜力（总体精度OA = 70–80%），而Sentinel-2的表现较低（OA = 53–73%）。冬季模型获得的精度评分高于春季模型，表明耕作判别存在季节性变化。研究结果凸显了高分辨率卫星数据结合机器学习在耕作方式制图和推进精准农业方面的实用性。","Journal of the Indian Society of Remote Sensing","2026-09-24T00:00:00Z",67,{"impact":21,"substance":82,"depth":83,"authority":84,"freshness":85,"relevant":22,"comment":86},20,17,13,9,"基于高分辨率卫星影像与随机森林识别耕作方式的实证研究，方法清晰、结论具体，对精准农业与耕地监测有参考价值，但属细分领域学术进展，公共影响有限。",[88],{"name":78,"url":75},[27,28,90,30,91],"保护性耕作","土壤耕作",[93,94],"PlanetScope Sentinel-2 耕作识别","保护性耕作 遥感 分类","PlanetScopeSentinel-2耕作识别-3499","10.1007\u002Fs12524-026-02580-1",{"doi":96,"openalex_id":98,"authors":99,"venue":78,"cited_by_count":36,"oa_url":75,"card":112,"direction":65,"ingested_from":67},"W7214193864",[100,103,106,109],{"name":101,"orcid":102},"Vidya Nahdhiyatul Fikriyah","https:\u002F\u002Forcid.org\u002F0000-0003-2869-3657",{"name":104,"orcid":105},"Roshanak Darvishzadeh","https:\u002F\u002Forcid.org\u002F0000-0001-7512-0574",{"name":107,"orcid":108},"Stephan M. Haefele","https:\u002F\u002Forcid.org\u002F0000-0003-0389-8373",{"name":110,"orcid":111},"Andrew Nelson","https:\u002F\u002Forcid.org\u002F0000-0002-7249-3778",{"tldr":113,"method":114,"finding":115,"direction":65,"opportunity":116},"对比PlanetScope与Sentinel-2影像，用随机森林区分冬春两季的集约与保护性耕作。","英国田间试验，2022-24年光谱时序数据，随机森林分类。","PlanetScope分类精度70-80%优于Sentinel-2，冬季模型精度高于春季，绿、红边和","可探索多源高分辨率影像融合与时序特征优化，提升不同季节和区域耕作分类的泛化能力。","2026-09-25T23:30:30.866355Z",{"id":119,"title":120,"url":121,"summary":122,"summary_zh":9,"content":9,"source_name":123,"source_url":9,"published_at":124,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":125,"score_detail":126,"sources":128,"tags":130,"search_phrases":133,"slug":136,"view_count":36,"doi":9,"paper":137,"created_at":146},3248,"Crop recommendation in precision agriculture: a systematic literature review of methods, trends, and challenges（精准农业中的作物推荐：方法、趋势与挑战系统综述）","https:\u002F\u002Fm2.mtmt.hu\u002Fapi\u002Fpublication\u002F37471110","MDPI 发表精准农业作物推荐方法系统综述：从183项研究中严格筛选129篇发表于2020-2026年的文章，使用PRISMA协议进行综合分析。研究表明集成学习方法（特别是随机森林和XGBoost）在各种农业数据集的预测性能上具有强大能力；支持向量机、决策树、k近邻等传统ML方法仍被广泛使用；同时CNN和LSTM被用于遥感和时间相关农业分析。最常用的数据集来源是Kaggle，典型输入包括土壤养分（NPK）、土壤pH、天气条件和NDVI、EVI等卫星指数。研究主要研究空白：有限的实时部署、低多数据源集成、低跨区域验证、低模型可解释性。研究表明可扩展、可解释的AI系统对农业实际应用具有重要意义。","MDPI","2026-09-22T00:00:00Z",81,{"impact":19,"substance":18,"depth":19,"authority":84,"freshness":13,"relevant":22,"comment":127},"基于PRISMA的129篇文献系统综述，梳理作物推荐主流方法与四大研究空白，对农业AI落地有参考价值。",[129],{"name":123,"url":121},[27,131,28,132,30],"农业人工智能","作物推荐",[134,135],"精准农业 作物推荐 系统综述","XGBoost 随机森林 作物推荐","精准农业作物推荐系统综述-3248",{"doi":9,"openalex_id":9,"authors":138,"venue":9,"cited_by_count":36,"oa_url":9,"card":139,"direction":143,"ingested_from":145},[],{"tldr":140,"method":141,"finding":142,"direction":143,"opportunity":144},"系统综述129篇2020-2026年文献，梳理精准农业作物推荐的方法、趋势与挑战。","PRISMA协议系统综述，分析183项研究筛选出的129篇文献。","集成学习（随机森林、XGBoost）表现最强，主要空白为实时部署、多源集成、跨区域验证与可解释性。","农业人工智能与决策模型","可探索可解释、可跨区域泛化的实时作物推荐系统，并融合多源遥感与物联网数据。","agent","2026-09-23T00:04:33.331160Z",{"id":148,"title":149,"url":150,"summary":151,"summary_zh":152,"content":9,"source_name":153,"source_url":150,"published_at":124,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":154,"score_detail":155,"sources":158,"tags":160,"search_phrases":163,"slug":166,"view_count":36,"doi":167,"paper":168,"created_at":192},3167,"Temporal trend analysis and multi-temporal satellite feature integration for mango orchard acreage estimation using machine learning algorithms approach","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-65828-3","Accurate estimation of horticultural crop acreage is essential for agricultural planning, market forecasting and evidence-based policy formulation. The present study investigated long-term trends in mango cultivation and developed a multi-temporal remote sensing framework for mango orchard acreage estimation in Navsari District using integrated optical, SAR and machine learning approaches. Time-series (temporal) data spanning 23 years (2001–02 to 2023–24) were analyzed using polynomial regression models to evaluate trends in mango area and production. Linear regression best represented area expansion trends (Adj. R² = 0.969), whereas cubic regression better captured production variability (Adj. R² = 0.634), indicating climatic and seasonal influences on productivity. For orchard classification and acreage estimation, multi-temporal Sentinel-2 imagery acquired from November 2023 to March 2024 was processed within a phenology-guided framework. Monthly composites were generated and integrated with Sentinel-1 SAR backscatter data, vegetation indices (NDVI, GNDVI, NDRE, SAVI, EVI and NDMI) and texture metrics derived from Gray Level Co-occurrence Matrix (GLCM) analysis. A comprehensive 72-band feature stack was developed for classification. Four machine learning algorithms, namely Random Forest (RF), XGBoost (XGB), Support Vector Machine (SVM) and Multinomial Logistic Regression (MNLR) were evaluated for orchard discrimination. Among the tested models, RF achieved the highest classification performance with an Overall Accuracy of 99.80% and a Kappa coefficient of 0.997, followed by SVM (99.30%), MNLR (98.21%) and XGB (9.20%). The RF model estimated mango orchard area at 36,099.94 ha, showing the closest agreement with official horticultural statistics (34,363 ha) with only 5.05% estimation error. In contrast, SVM and MNLR overestimated orchard extent by 16.03% and 43.37%, respectively. The proposed framework provides a reliable and scalable methodology for operational horticultural monitoring, crop inventory generation and precision agricultural planning in tropical orchard ecosystems.","准确估算园艺作物种植面积对于农业规划、市场预测和循证政策制定至关重要。本研究探讨了芒果种植的长期趋势，并开发了一个多时相遥感框架，结合光学、合成孔径雷达（SAR）和机器学习方法，用于纳夫萨里县芒果园种植面积估算。利用多项式回归模型分析了跨越23年（2001—02年至2023—24年）的时间序列数据，以评估芒果面积和产量的变化趋势。线性回归最能表征面积扩张趋势（调整R² = 0.969），而三次回归更能捕捉产量变异性（调整R² = 0.634），表明气候和季节性因素对生产力具有影响。在果园分类和面积估算方面，基于物候指导框架处理了2023年11月至2024年3月获取的多时相Sentinel-2影像。生成了月度合成影像，并将其与Sentinel-1 SAR后向散射数据、植被指数（NDVI、GNDVI、NDRE、SAVI、EVI和NDMI）以及基于灰度共生矩阵（GLCM）分析提取的纹理指标进行整合。构建了一个包含72个波段的综合特征集用于分类。评估了四种机器学习算法，即随机森林（RF）、XGBoost（XGB）、支持向量机（SVM）和多项逻辑回归（MNLR），用于果园判别。在测试的模型中，RF取得了最高的分类性能，总体精度为99.80%，Kappa系数为0.997，其次是SVM（99.30%）、MNLR（98.21%）和XGB（9.20%）。RF模型估算的芒果园面积为36,099.94公顷，与官方园艺统计数据（34,363公顷）最为接近，估算误差仅为5.05%。相比之下，SVM和MNLR分别高估了果园面积16.03%和43.37%。所提出的框架为热带果园生态系统中的业务化园艺监测、作物清单生成和精准农业规划提供了一种可靠且可扩展的方法。","Scientific Reports",78,{"impact":156,"substance":18,"depth":19,"authority":156,"freshness":13,"relevant":22,"comment":157},14,"方法扎实、数据规模大且精度高，但属区域性作物遥感估产研究，产业影响有限，可作为技术方法类精选。",[159],{"name":153,"url":150},[27,28,161,30,162],"芒果","作物估产",[164,165],"Navsari 芒果 遥感估产","Sentinel-2 芒果 果园面积","Navsari芒果遥感估产-3167","10.1038\u002Fs41598-026-65828-3",{"doi":167,"openalex_id":169,"authors":170,"venue":153,"cited_by_count":36,"oa_url":150,"card":187,"direction":65,"ingested_from":67},"W7213920056",[171,173,176,178,180,183,185],{"name":172,"orcid":9},"V. Raju",{"name":174,"orcid":175},"Yogesh A. Garde","https:\u002F\u002Forcid.org\u002F0000-0002-0297-316X",{"name":177,"orcid":9},"Dr. V. S. Thorat",{"name":179,"orcid":9},"V. T. Shinde",{"name":181,"orcid":182},"Nitin Varshney","https:\u002F\u002Forcid.org\u002F0000-0001-9144-5475",{"name":184,"orcid":9},"Alok Shrivastava",{"name":186,"orcid":9},"A. P. Chaudhary",{"tldr":188,"method":189,"finding":190,"direction":65,"opportunity":191},"融合多时相Sentinel-1\u002F2与机器学习，估算印度芒果园面积并分析23年种植趋势。","23年时序回归分析；Sentinel-2月合成+SAR+植被指数+GLCM纹理共","RF精度最高（总体精度99.80%，Kappa 0.997），面积估算误差仅5.05%，优于SVM和","可迁移该多时相SAR-光学特征框架至其他热带果园，并探索深度学习与物候自适应特征优化。","2026-09-22T23:30:22.619289Z",{"id":194,"title":195,"url":196,"summary":197,"summary_zh":198,"content":9,"source_name":199,"source_url":196,"published_at":200,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":201,"score_detail":202,"sources":204,"tags":206,"search_phrases":209,"slug":212,"view_count":22,"doi":213,"paper":214,"created_at":243},2536,"Mapping Native Grass Cover with Random Forest Models: Sentinel-2 Versus Sentinel-2 Combined with Sentinel-1 SAR-Derived GLCM Texture Metrics","https:\u002F\u002Fdoi.org\u002F10.3390\u002Frs18183150","Temperate native grasslands in southeastern Australia have been extensively cleared for agriculture, and the remaining patches are under growing pressure from further land use change, climate variability, and invasive species. Mapping and monitoring their distribution and the cover of native and exotic grasses are critical for their conservation and management. Field-based methods are not always scalable or time-effective, and this study aimed to develop a scalable method to map and monitor the fractional cover-class maps of native C3 and native C4 grass cover as a component of remnant native grasslands on the western outskirts of Melbourne, Victoria, Australia. Field-based reference data for training and validation of random forest machine learning models were collected across multiple sites in 2021. Sentinel-2 optical spectral bands and vegetation indices were used as the primary input data, and Sentinel-1 Synthetic Aperture Radar (SAR)-derived Grey Level Co-occurrence Matrix (GLCM) texture metrics were assessed for their capacity to improve the model. Results show that random forest models trained on Sentinel-2 data without GLCM texture information derived from Sentinel-1 SAR data provided a moderate overall accuracy (C3: 59.1%, C4: 78.1%). Class-specific metrics showed that reliability was highest for better represented lower-cover classes, particularly the 6–25% native C3 class and the 0–5% native C4 class, while higher-cover classes were less reliable because of the limited number of training and validation samples. Grass cover fractions were modelled well for sparse to moderate grass cover, but dense grass cover was not modelled accurately, probably due to limited high-cover samples in the training dataset. Model performance was not improved by the inclusion of Sentinel-1 SAR-derived GLCM texture metrics, indicating that C-band VH-polarised SAR is not sensitive to the fine-scale structural heterogeneity that characterises native grassland ecosystems. Sparse native C3 and C4 grasses could be mapped most reliably in the lower-cover classes as a component of grasslands with optical remote sensing, and the method developed here can now be applied to enable evidence-based management of grasslands, biodiversity conservation and the monitoring of grassland composition in the WGR and elsewhere. Higher-resolution structural datasets and more sophisticated machine learning approaches may be required to accurately predict native C3 and C4 grass cover fractions in denser grasslands.","澳大利亚东南部的温带原生草原已被大面积开垦用于农业，残余斑块正面临土地利用进一步变化、气候变率和入侵物种日益增大的压力。对其分布以及原生和外来草类覆盖度进行制图和监测，对于草原的保护和管理至关重要。基于实地调查的方法并非总是可扩展或省时的，本研究旨在开发一种可扩展的方法，以制图和监测作为残余原生草原组成部分的原生C3和原生C4草类覆盖度的分数覆盖等级图，研究区位于澳大利亚维多利亚州墨尔本西郊。用于训练和验证随机森林机器学习模型的实地参考数据于2021年在多个样点采集。研究以Sentinel-2光学光谱波段和植被指数作为主要输入数据，并评估了Sentinel-1合成孔径雷达（SAR）衍生的灰度共生矩阵（GLCM）纹理指标对提升模型性能的能力。结果表明，仅使用Sentinel-2数据（不含Sentinel-1 SAR衍生的GLCM纹理信息）训练的随机森林模型提供了中等的总体精度（C3：59.1%，C4：78.1%）。分类别指标显示，对于代表性较好的低覆盖度类别，可靠性最高，尤其是6–25%原生C3类别和0–5%原生C4类别，而较高覆盖度类别的可靠性较低，原因是训练和验证样本数量有限。草类覆盖度分数在稀疏至中等草类覆盖条件下建模效果良好，但茂密草类覆盖未能准确建模，可能是由于训练数据集中高覆盖度样本有限。纳入Sentinel-1 SAR衍生的GLCM纹理指标并未改善模型性能，表明C波段VH极化SAR对原生草原生态系统所特有的精细尺度结构异质性不敏感。作为草原的组成部分，稀疏的原生C3和C4草类在低覆盖度类别中利用光学遥感可最可靠地制图，本研究开发的方法现可应用于西维多利亚草原（WGR）及其他地区，以实现基于证据的草原管理、生物多样性保护和草原组成监测。准确预测原生C3和C4草类覆盖度分数可能需要更高分辨率的结构数据集和更先进的机器学习方法。","Remote Sensing","2026-09-13T00:00:00Z",71,{"impact":20,"substance":82,"depth":83,"authority":156,"freshness":21,"relevant":22,"comment":203},"方法新颖、结论明确（SAR纹理未提升精度），对草地遥感监测有参考价值，但属细分领域研究，影响范围有限。",[205],{"name":199,"url":196},[27,28,30,207,208],"草原生态","植被覆盖",[210,211],"智慧农业 机器学习 植被覆盖 草原生态","智慧农业 机器学习","智慧农业机器学习植被覆盖草原生态-2536","10.3390\u002Frs18183150",{"doi":213,"openalex_id":215,"authors":216,"venue":199,"cited_by_count":36,"oa_url":196,"card":238,"direction":65,"ingested_from":67},"W7212561645",[217,220,223,226,228,230,233,235],{"name":218,"orcid":219},"Sabah Sabaghy","https:\u002F\u002Forcid.org\u002F0000-0002-9453-8922",{"name":221,"orcid":222},"M. Abuzar","https:\u002F\u002Forcid.org\u002F0000-0002-6101-1307",{"name":224,"orcid":225},"Steve J. Sinclair","https:\u002F\u002Forcid.org\u002F0000-0002-4282-1021",{"name":227,"orcid":9},"Tony Dugdale",{"name":229,"orcid":9},"Vanessa Hutchins",{"name":231,"orcid":232},"Yogendra K. Karna","https:\u002F\u002Forcid.org\u002F0000-0002-2120-4710",{"name":234,"orcid":9},"Jonathan Wilson",{"name":236,"orcid":237},"Kathryn Sheffield","https:\u002F\u002Forcid.org\u002F0000-0003-2624-9739",{"tldr":239,"method":240,"finding":241,"direction":65,"opportunity":242},"用随机森林结合Sentinel-2与Sentinel-1纹理特征，绘制澳洲原生草地C3\u002FC4草覆盖度","2021年野外样点训练随机森林，Sentinel-2光谱与植被指数为主，Sent","仅用Sentinel-2精度中等（C3 59.1%、C4 78.1%），加入SAR纹理未提升，高覆盖","高覆盖度草地样本不足且C波段SAR不敏感，可探索高分辨率结构数据与深度模型提升密草覆盖反演。","2026-09-15T23:30:21.287053Z",{"id":245,"title":246,"url":247,"summary":248,"summary_zh":249,"content":9,"source_name":250,"source_url":247,"published_at":251,"category":12,"cover_url":9,"hotness":252,"is_selected":14,"score":253,"score_detail":254,"sources":258,"tags":262,"search_phrases":263,"slug":266,"view_count":36,"doi":267,"paper":268,"created_at":281},1527,"Smart Crop Recommendation System","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22245870","Agriculture remains the primary driver of India's economic stability and food security. However, conventional farming heavily relies on subjective experience rather than scientific data, resulting in sub-optimal crop selection, improper resource usage, and heightened vulnerability to changing weather and plant diseases. To resolve these limitations, this paper presents the Smart Crop Recommendation System, an integrated decision-support web platform utilizing Artificial Intelligence, Machine Learning, Deep Learning, and Real-time Weather Analytics. The proposed application evaluates soil composition—specifically Nitrogen (N), Phosphorus (P), Potassium (K), and pH levels—alongside environmental parameters including temperature, humidity, and rainfall to accurately recommend optimal crops using Scikit-Learn classification algorithms. Real-time weather forecasting is integrated via the OpenWeather API to guide critical agricultural schedules such as sowing and irrigation. Furthermore, a Deep Learning module employing a Convolutional Neural Network (CNN) detects crop diseases from uploaded leaf images and outputs targeted treatment strategies. Implemented with a Django web framework, SQLite database, Power BI analytical dashboards, and cloud infrastructure, the platform offers an end-to-end digital assistant that boosts yield productivity, minimizes farming risks, and supports sustainable precision agriculture.","农业仍然是印度经济稳定和粮食安全的主要驱动力。然而，传统农业严重依赖主观经验而非科学数据，导致作物选择欠佳、资源利用不当，以及对气候变化和植物病害的脆弱性增加。为解决这些局限性，本文提出了智能作物推荐系统——一个集成的决策支持网络平台，利用人工智能、机器学习、深度学习和实时天气分析技术。该应用系统评估土壤成分——特别是氮（N）、磷（P）、钾（K）和pH值——以及包括温度、湿度和降雨量在内的环境参数，通过Scikit-Learn分类算法准确推荐最佳作物。通过OpenWeather API集成实时天气预报，以指导播种和灌溉等关键农事安排。此外，采用卷积神经网络（CNN）的深度学习模块可从上传的叶片图像中检测作物病害，并输出针对性的治理策略。该平台基于Django网络框架、SQLite数据库、Power BI分析仪表板和云基础设施实现，提供端到端的数字助手，可提高产量生产力、降低农业风险，并支持可持续的精准农业。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-02T00:00:00Z",25,62,{"impact":20,"substance":19,"depth":255,"authority":13,"freshness":256,"relevant":22,"comment":257},15,7,"系统整合AI与气象数据，提供作物推荐与病害检测，对精准农业有参考价值。",[259,260],{"name":250,"url":247},{"name":250,"url":261},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22245869",[27,131,28,132,31],[264,265],"农业人工智能 作物推荐 智慧农业 机器学习","农业人工智能 作物推荐","农业人工智能作物推荐智慧农业机器学习-1527","10.5281\u002Fzenodo.22245870",{"doi":267,"openalex_id":269,"authors":270,"venue":250,"cited_by_count":36,"oa_url":247,"card":275,"direction":280,"ingested_from":67},"W7206166550",[271,273],{"name":272,"orcid":9},"Jayashree S P",{"name":274,"orcid":9},"S Sahana",{"tldr":276,"method":277,"finding":278,"direction":143,"opportunity":279},"提出智能作物推荐系统，结合AI与实时天气，推荐作物并检测病害。","使用Scikit-Learn分类算法、CNN、OpenWeather API、D","系统能提高产量、降低风险，支持可持续精准农业。","可扩展至多作物区域适应性、考虑经济因素及用户反馈的个性化推荐。","智慧农业 \u002F 农业物联网","2026-09-03T23:30:09.661533Z",{"id":283,"title":284,"url":285,"summary":286,"summary_zh":287,"content":9,"source_name":288,"source_url":285,"published_at":79,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":154,"score_detail":289,"sources":291,"tags":293,"search_phrases":296,"slug":299,"view_count":36,"doi":300,"paper":301,"created_at":314},3517,"A Resource-Efficient Hybrid CNN-LSTM Network for Image-Based Bean Leaf Disease Classification","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fjimaging12100468","Accurate and resource-efficient automated diagnosis is a cornerstone of modern agricultural expert systems. While Convolutional Neural Networks (CNNs) have established benchmarks in plant pathology, their ability to capture long-range spatial dependencies is often limited by standard pooling layers, and their high memory footprint hinders deployment on portable devices. This paper proposes a lightweight hybrid CNN-LSTM system for bean leaf disease classification. By integrating an LSTM layer to model the spatial–sequential relationships within feature maps, our hybrid architecture achieves a 94.36% accuracy and 94.38% F1 score while maintaining an exceptionally small footprint of 1.86 MB, a 70% reduction in size compared to traditional CNN-based systems. Furthermore, we provide a systematic evaluation of image augmentation strategies, demonstrating that tailored transformations are superior to generic combinations for maintaining the integrity of diagnostic patterns. Results on the ibean dataset confirm that the proposed system achieves state-of-the-art F1 scores of 99.22% with EfficientNet-B7+LSTM, providing a potentially robust and scalable framework for real-time agricultural decision support in resource-constrained environments. The code and augmented datasets used in this study are publicly available on this GitHub repo.","准确且资源高效的自动化诊断是现代农业专家系统的基石。尽管卷积神经网络（CNN）在植物病理学领域已确立了基准，但其捕捉长程空间依赖关系的能力常受限于标准池化层，且高内存占用阻碍了其在便携设备上的部署。本文提出了一种用于豆叶病害分类的轻量级混合CNN-LSTM系统。通过集成LSTM层来建模特征图内的空间-序列关系，我们的混合架构达到了94.36%的准确率和94.38%的F1分数，同时保持了仅1.86 MB的极小占用，相较于传统基于CNN的系统体积减少了70%。此外，我们系统评估了图像增强策略，表明定制化变换在保持诊断模式完整性方面优于通用组合。在ibean数据集上的结果证实，所提出的系统结合EfficientNet-B7+LSTM达到了99.22%的最先进F1分数，为资源受限环境中的实时农业决策支持提供了一个潜在稳健且可扩展的框架。本研究使用的代码和增强数据集已在此GitHub仓库公开。","Journal of Imaging",{"impact":17,"substance":18,"depth":19,"authority":84,"freshness":85,"relevant":22,"comment":290},"轻量级CNN-LSTM豆叶病害识别研究，方法新颖、数据可靠且代码开源，对资源受限场景下的农业智能诊断有实用价值。",[292],{"name":288,"url":285},[27,131,31,294,295],"轻量化模型","豆类作物",[297,298],"ibean dataset 豆叶病害","农业人工智能 轻量化模型 智慧农业 病害识别","ibeandataset豆叶病害-3517","10.3390\u002Fjimaging12100468",{"doi":300,"openalex_id":302,"authors":303,"venue":288,"cited_by_count":36,"oa_url":285,"card":309,"direction":143,"ingested_from":67},"W7154572440",[304,306],{"name":305,"orcid":9},"Hye Jin Rhee",{"name":307,"orcid":308},"Joseph Damilola Akinyemi","https:\u002F\u002Forcid.org\u002F0000-0003-3121-4231",{"tldr":310,"method":311,"finding":312,"direction":143,"opportunity":313},"提出轻量级CNN-LSTM混合网络，用于豆叶病害分类，兼顾高精度与低资源占用。","CNN提取特征后接LSTM建模空间序列关系，在ibean数据集上评估并系统比较图","模型准确率94.36%、F1 94.38%，仅1.86MB，比传统CNN缩小70%，Efficien","可探索面向移动端\u002F边缘设备的超轻量病害诊断模型，并研究增强策略与模型结构的自适应协同优化。","2026-09-25T23:30:59.276381Z"]