[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3013":3,"related-3013":61},{"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":60},3013,"AI and remote sensing for fungal and oomycete disease surveillance: current landscape and biological integration","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs41348-026-01352-w","AI and remote sensing for fungal and oomycete disease surveillance: current landscape and biological integration。Journal of Plant Diseases and Protection","人工智能与遥感在真菌及卵菌病害监测中的应用：现状与生物学整合。《植物病害与保护杂志》",null,"Journal of Plant Diseases and Protection","2026-09-19T00:00:00Z","论文",10,false,77,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},18,20,17,13,9,1,"核心期刊综述，系统梳理AI与遥感在真菌及卵菌病害监测中的进展与生物学整合路径，对智慧农业植保方向有参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","植物病害","遥感监测","病害预警",[33,34],"AI 遥感 真菌病害 监测","植物病害 遥感 预警","AI遥感真菌病害监测-3013",0,"10.1007\u002Fs41348-026-01352-w",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":9,"card":53,"direction":57,"ingested_from":59},"W7213649225",[41,43,45,47,50],{"name":42,"orcid":9},"Biju Vadakkemukadiyil Chellappan",{"name":44,"orcid":9},"C. L. Biji",{"name":46,"orcid":9},"Vanshika Arun Meda",{"name":48,"orcid":49},"Sajad Ali","https:\u002F\u002Forcid.org\u002F0000-0002-3230-1436",{"name":51,"orcid":52},"Sherif Mohamed El‐Ganainy","https:\u002F\u002Forcid.org\u002F0000-0001-5226-4604",{"tldr":54,"method":55,"finding":56,"direction":57,"opportunity":58},"综述AI与遥感在真菌及卵菌病害监测中的现状，强调生物信息整合。","文献综述，整合AI、遥感与病原生物学数据。","AI与遥感结合可提升病害监测，但需融入病原生物学机制。","农业遥感与作物表型","可研究将病原生活史与遥感时序特征耦合的病害预警模型。","openalex","2026-09-20T23:30:21.177583Z",{"total":62,"page":22,"page_size":62,"items":63},6,[64,90,141,195,226,251],{"id":65,"title":66,"url":67,"summary":68,"summary_zh":9,"content":69,"source_name":70,"source_url":9,"published_at":71,"category":72,"cover_url":9,"hotness":13,"is_selected":14,"score":73,"score_detail":74,"sources":79,"tags":81,"search_phrases":85,"slug":88,"view_count":36,"doi":9,"paper":9,"created_at":89},2898,"苏垦农发神农慧种农业AI大模型规模化落地:天空地一体化闭环,百万亩自有农田实景数据","https:\u002F\u002Fcaifuhao.eastmoney.com\u002Fnews\u002F20260918101757264727920","苏垦农发9月18日发文,公司依托百万亩自有连片高标准农田,持续产出真实大田数据训练神农慧种农业AI智能体,实现天空地一体化数据闭环:空中多光谱无人机集群常态化农田巡测;地面全域四情监测传感器、北斗智能农机、智能灌溉终端;云端苏垦智云平台+神农慧种AI模型,形成采集数据→AI分析研判→输出水肥植保方案→农机落地执行完整闭环。苏垦智云是全国农林牧渔领域唯一入选工信部信创典型案例的农业数字化平台。","[在东方财富看资讯行情，选东方财富证券一站式开户交易>>](https:\u002F\u002Facttg.eastmoney.com\u002Fpub\u002Fwebtg_hskh_act_zixun_01_01_01_0)\n\n（国内A股找不到第二家，像苏垦农发依托百万亩自有连片高标准农田，持续产出真实大田数据训练神农慧种农业AI智能体；苏垦实现天空地一体化数据闭环，苏垦智云是农林牧渔唯一工信部信创典型案例，智慧农业+低空经济双主线落地。）\n\n- 空中：多光谱无人机集群开展农田巡测；\n\n![Image 1](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002FB9B393E75E9CCC55E11A2686662765C9_w1080h720.jpg)\n\n- 地面：农田四情监测传感器、北斗智能农机；\n\n![Image 2](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F4D780511671CDAB50B4A1CCFE057703D_w1339h892.jpg)\n\n- 云端：苏垦智云平台与神农慧种AI模型，形成「采集数据→AI分析研判→输出水肥植保方案→农机落地执行」完整闭环。\n\n![Image 3](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F1059027DFA2A605F6861D7C69D6D4309_w1440h1080.jpg)\n\n百万亩自有连片农田源源不断产出真实田间数据，持续迭代优化AI模型。国内很多农业AI企业仅拥有小片试验田，唯有苏垦农发拥有大规模现代农业实景数据用于农业模型训练。\n\n苏垦智云平台，也是全国农林牧渔领域唯一入选工信部信创典型案例的农业数字化平台。苏垦农发一一智慧农业与低空经济的天空地一体化闭环落地，AI大田规模化实体应用！\n\n苏垦农发打造天空地一体化智慧农业完整体系，AI大田并非实验室试验，而是在百万亩自有连片农田实现规模化落地运行。\n\n1、空中：多光谱无人机集群常态化农田巡测，低空遥感采集作物长势、病虫害、墒情数据；依托农业农村部低空技术创新重点实验室，主攻低空多模态农情感知。\n\n2、地面：全域农田“四情”监测传感器、北斗智能农机、智能灌溉终端，实时采集土壤、苗情、虫情、气象数据。\n\n3、云端：苏垦智云平台 神农慧种农业AI智能体，形成完整闭环：采集田间数据→AI模型分析研判→输出水肥、植保作业处方→下发农机执行落地。\n\n4、核心稀缺壁垒：手握百万亩自有连片高标准农田，源源不断产出真实大田实景数据，持续迭代训练神农慧种AI模型。\n\n国内绝大多数农业AI企业，仅拥有小片试验田做演示；苏垦是少数拥有大规模真实农业场景用于模型训练与生产验证的实体龙头。\n\n5、苏垦智云一体化平台，也是全国农林牧渔领域唯一入选工信部信创典型案例的农业数字化平台，国产化底层架构，是农业数字化可复制的标杆样板。配套全国首个农业农村部农业低空技术创新重点实验室（苏垦为依托单位、河海大学共建），同步布局低空经济与智慧农业新质生产力。\n\n2026-09-18 11:16:07 作者更新了以下内容\n\n全球领先的风险咨询公司Verisk Maplecroft 在周四（9月17日）发布的一份报告中表示，极端天气灾害将加剧亚洲的粮食安全风险，并可能在印度、印尼和菲律宾等脆弱的国家引发动荡。\n\n![Image 4](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F9ACF9C92E7B35A242970CDC4D55B8DA9_w1080h15645.jpg)\n\n![Image 5](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002FFE1BF075831D4DAEB94ADAE924123EE9_w1080h2400.jpg)\n\n2026-09-18 21:02:06 作者更新了以下内容\n\n苏垦农发一一 AI赋能农业真实落地案例：临海农场——国内首个10万亩级无人值守巡田农场（核心标杆）\n\n地点：江苏盐城临海农场，苏垦智慧农业科技园\n\n1. 空中低空遥感AI巡田\n\n![Image 6](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002FE4E4DDE3ED1EEF061A2D631622A46F73_w1424h800.jpg)\n\n多光谱无人机集群常态化巡航，采集苗情、墒情、病虫害影像数据，AI自动识别长势差异、病斑，生成热力图；替代人工徒步巡田，十几分钟就能完成万亩农田普查。依托农业农村部低空技术创新重点实验室，开展低空多模态农情感知研究。\n\n2. AI智能光伏远程灌溉系统\n\n![Image 7](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F9FF2DAAEBED53D7BB8C825B388C102BD_w1424h800.jpg)\n\n万亩稻田布设太阳能智能闸门，通过土壤墒情传感器采集数据，AI分析土壤缺水程度，手机APP一键远程开关水渠闸门。\n\n量化效果：过去管500亩农田，人工开关闸门半天；现在2分钟完成全部闸门调控，灌溉效率提升20倍，每亩节约管水人工成本约30元。\n\n3. AR眼镜AI虫害识别\n\n![Image 8](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F9C7DFBF3F14BDE7E73309CA953C0C8E7_w1424h800.jpg)\n\n农技人员佩戴AR眼镜在田间巡查，拍摄虫体，AI毫秒级识别稻飞虱等害虫种类、统计虫口密度，识别准确率＞95%，自动推送防治方案，新手农技员也能快速判别田间虫害。\n\n4. AI变量施肥无人机作业\n\n![Image 9](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002FC257CCA1B1E8AE1621C403E96DF222EE_w1424h800.jpg)\n\nAI读取水稻营养、长势数据，为每一块条田生成独立追肥处方，无人机分区精准施肥，一地一策，实现肥药双减，农药化肥年均用量下降约3%。\n\n5. 北斗智能农机 AI收割决策\n\n![Image 10](https:\u002F\u002Fgbres.dfcfw.com\u002FFiles\u002Fpicture\u002F20260918\u002F00A2AE87C3E170BFBFE610FABE2556A5_w1424h800.jpg)\n\n北斗导航插秧机、无人收割机，AI根据成熟度、含水率数据，指导分块错峰收割，减少粮食收割损耗。\n\n[恭喜解锁12个月手机L2专属领取资格，立即领取>>](https:\u002F\u002Facttg.eastmoney.com\u002Fpub\u002Fwebtg_hskh_act_zixun_01_01_01_0)\n\n暗盘资金榜已更新!这些个股\u002F板块可以关注>\n\n追加内容\n\n本文作者可以追加内容哦 !\n\n**郑重声明：**用户在社区发表的所有信息将由本网站记录保存，仅代表作者个人观点，与本网站立场无关，不对您构成任何投资建议，据此操作风险自担。**请勿相信代客理财、免费荐股和炒股培训等宣传内容，远离非法证券活动。请勿添加发言用户的手机号码、公众号、微博、微信及QQ等信息，谨防上当受骗！**\n\n[![Image 11](https:\u002F\u002Favator.eastmoney.com\u002Fqface\u002F9825094237066000\u002F360)](https:\u002F\u002Fi.eastmoney.com\u002F9825094237066000)\n\n总收益 20日收益 日收益\n------\n\n历史收益率走势(%)\n\nChart\n\n代码 名称 最新价 涨跌幅\n[查看更多](http:\u002F\u002Figuba.eastmoney.com\u002F9825094237066000)\n\n浪客视频\n\n![Image 12](https:\u002F\u002Fnp-newspic.dfcfw.com\u002Fdownload\u002FD25261481966621695940_w340h340.jpg)\n\n![Image 13](https:\u002F\u002Fgbapi.eastmoney.com\u002Fshareopt\u002Fweb\u002Fweb_click.gif?id=20260918101757264727920&type=20&version=200&product=EastMoney&plat=Web&deviceid=caifuhao)\n\n郑重声明：东方财富网发布此信息的目的在于传播更多信息，与本站立场无关。东方财富网不保证该信息（包括但不限于文字、视频、音频、数据及图表）全部或者部分内容的准确性、真实性、完整性、有效性、及时性、原创性等。相关信息并未经过本网站证实，不对您构成任何投资建议，据此操作，风险自担。","东方财富财富号\u002F苏垦农发","2026-09-18T00:00:00Z","报道",69,{"impact":75,"substance":17,"depth":76,"authority":77,"freshness":13,"relevant":22,"comment":78},22,14,5,"苏垦农发百万亩自有农田上实现天空地一体化AI闭环，含临海农场10万亩无人巡田等量化案例，产业参考价值较高，但来源为财富号自媒体、宣传色彩浓，权威性偏弱。",[80],{"name":70,"url":67},[27,82,28,83,30,84],"低空经济","智能农机","数字农田",[86,87],"苏垦农发 神农慧种 AI大模型","临海农场 无人值守巡田","苏垦农发神农慧种AI大模型-2898","2026-09-19T00:06:07.612319Z",{"id":91,"title":92,"url":93,"summary":94,"summary_zh":95,"content":9,"source_name":96,"source_url":93,"published_at":97,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":98,"sources":100,"tags":102,"search_phrases":105,"slug":108,"view_count":36,"doi":109,"paper":110,"created_at":140},2801,"A comprehensive survey on machine vision applications in precision agriculture: current trends and future perspectives","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs41060-026-01278-4","A comprehensive survey on machine vision applications in precision agriculture: current trends and future perspectives。International Journal of Data Science and Analytics","精准农业中机器视觉应用的综合综述：当前趋势与未来展望。《国际数据科学与分析杂志》","International Journal of Data Science and Analytics","2026-09-17T00:00:00Z",{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":99},"核心期刊发表的机器视觉精准农业综述，方法梳理与趋势判断具参考价值，但属综述类论文，产业影响有限。",[101],{"name":96,"url":93},[27,28,103,30,104],"精准农业","机器视觉",[106,107],"农业人工智能 智慧农业 机器视觉 精准农业","农业人工智能 智慧农业","农业人工智能智慧农业机器视觉精准农业-2801","10.1007\u002Fs41060-026-01278-4",{"doi":109,"openalex_id":111,"authors":112,"venue":96,"cited_by_count":36,"oa_url":9,"card":134,"direction":138,"ingested_from":59},"W7213471057",[113,115,117,119,122,124,126,129,132],{"name":114,"orcid":9},"Shirun Gu",{"name":116,"orcid":9},"Xinyuan Fan",{"name":118,"orcid":9},"Lihui Zhu",{"name":120,"orcid":121},"Caixia Song","https:\u002F\u002Forcid.org\u002F0000-0003-3897-7629",{"name":123,"orcid":9},"Lei Mu",{"name":125,"orcid":9},"Zichen Zhang",{"name":127,"orcid":128},"Rui Zhang","https:\u002F\u002Forcid.org\u002F0000-0002-8634-3519",{"name":130,"orcid":131},"Tong Xu","https:\u002F\u002Forcid.org\u002F0000-0001-5564-192X",{"name":133,"orcid":9},"Zhiyuan Zhang",{"tldr":135,"method":136,"finding":137,"direction":138,"opportunity":139},"综述机器视觉在精准农业中的应用现状与未来趋势。","文献综述，梳理机器视觉在精准农业中的技术路线。","机器视觉已广泛用于作物监测、病虫害识别等，但落地仍受数据与算力限制。","农业人工智能与决策模型","可聚焦轻量化模型与边缘部署，解决田间实时性与数据稀缺问题。","2026-09-17T23:30:54.103781Z",{"id":142,"title":143,"url":144,"summary":145,"summary_zh":146,"content":9,"source_name":147,"source_url":144,"published_at":148,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":149,"score_detail":150,"sources":154,"tags":156,"search_phrases":159,"slug":162,"view_count":36,"doi":163,"paper":164,"created_at":194},2670,"A knowledge-guided machine learning framework for cross-scale wheat harvest monitoring via sample augmentation","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.rse.2026.115671","Accurate monitoring of wheat harvest is crucial for precision agriculture and ensuring food security. However, rapid changes in land surface composition during the harvest period in intensive agricultural regions make it difficult to obtain sufficiently high-confidence ground samples, limiting the performance and generalization of data-driven remote sensing methods. Therefore, this study proposes a Knowledge-Guided Machine Learning (KGML) framework that integrates multi-satellite Earth observation data (PlanetScope, Sentinel-2, and MODIS) to monitor harvest from the field to regional scales. Ground data were collected using vehicle-mounted cameras and smartphones during the 2023 and 2024 wheat harvest periods. The results showed that combining spectral knowledge rules with a Random Forest model (regional accuracy >0.80) generated numerous high-confidence augmented samples from PlanetScope imagery. The augmented dataset was used to train a Hybrid CNN-Transformer-LSTM (HCTL) model with two pathways: Sentinel-2 classification for field-level harvest mapping (overall accuracy = 0.93) and MODIS regression for sub-pixel harvest fraction estimation, which showed high agreement with PlanetScope-derived harvest fractions (R 2 = 0.97, RMSE = 0.07, rRMSE = 0.15). The harvest dates derived from the MODIS harvest fraction time series showed high consistency with field observations (R 2 = 0.82, RMSE = 1.30 days). This framework provides an effective solution for wheat harvest monitoring by bridging the gap between limited ground-truth data and multi-scale satellite observations, thereby supporting food security assessments and informed agricultural management decisions.","准确监测小麦收获对精准农业和保障粮食安全至关重要。然而，在集约化农业区域，收获期地表组成的快速变化使得获取足够高置信度的地面样本变得困难，限制了数据驱动遥感方法的性能和泛化能力。因此，本研究提出了一种知识引导机器学习（KGML）框架，集成多卫星地球观测数据（PlanetScope、Sentinel-2和MODIS），实现从田块到区域尺度的收获监测。地面数据通过车载摄像头和智能手机在2023年和2024年小麦收获期采集。结果表明，将光谱知识规则与随机森林模型相结合（区域精度>0.80），可从PlanetScope影像中生成大量高置信度增强样本。利用该增强数据集训练了混合CNN-Transformer-LSTM（HCTL）模型，该模型包含两条路径：Sentinel-2分类用于田块尺度收获制图（总体精度=0.93），MODIS回归用于亚像元收获比例估算，其结果与PlanetScope-derived收获比例高度一致（R²=0.97，RMSE=0.07，rRMSE=0.15）。由MODIS收获比例时间序列提取的收获日期与田间观测结果高度一致（R²=0.82，RMSE=1.30天）。该框架通过弥合有限地面真值数据与多尺度卫星观测之间的差距，为小麦收获监测提供了有效解决方案，从而支持粮食安全评估和农业管理决策。","Remote Sensing of Environment","2026-09-15T00:00:00Z",87,{"impact":75,"substance":151,"depth":17,"authority":152,"freshness":21,"relevant":22,"comment":153},23,15,"提出知识引导机器学习框架，融合多源卫星数据实现田块到区域尺度的跨尺度小麦收获监测，方法新颖、精度可靠，对精准农业与粮食安全评估有实质参考价值。",[155],{"name":147,"url":144},[27,28,157,30,158],"粮食安全","小麦收获",[160,161],"农业人工智能 小麦收获 智慧农业 粮食安全","农业人工智能 小麦收获","农业人工智能小麦收获智慧农业粮食安全-2670","10.1016\u002Fj.rse.2026.115671",{"doi":163,"openalex_id":165,"authors":166,"venue":147,"cited_by_count":36,"oa_url":144,"card":189,"direction":57,"ingested_from":59},"W7213296259",[167,170,172,174,176,178,180,182,185,187],{"name":168,"orcid":169},"Mingchao Shao","https:\u002F\u002Forcid.org\u002F0000-0003-2619-4272",{"name":171,"orcid":9},"Chongya Jiang",{"name":173,"orcid":9},"Jingwei An",{"name":175,"orcid":9},"Haokai Zhu",{"name":177,"orcid":9},"Yue Li",{"name":179,"orcid":9},"Xia Yao",{"name":181,"orcid":9},"Tao Cheng",{"name":183,"orcid":184},"Hengbiao Zheng","https:\u002F\u002Forcid.org\u002F0009-0008-4778-0450",{"name":186,"orcid":9},"Weixing Cao",{"name":188,"orcid":9},"Yan Zhu",{"tldr":190,"method":191,"finding":192,"direction":57,"opportunity":193},"提出知识引导机器学习框架，用样本增强实现田块到区域尺度的冬小麦收获监测。","融合PlanetScope、Sentinel-2、MODIS与车载相机地面数据，","增强样本训练的HCTL模型田块分类精度0.93，区域收获比例R²=0.97，收获日期误差约1.3天。","可迁移至其他作物收获监测，并探索知识规则自动化构建与跨区域泛化能力。","2026-09-16T23:30:30.474537Z",{"id":196,"title":197,"url":198,"summary":199,"summary_zh":9,"content":9,"source_name":200,"source_url":9,"published_at":201,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":202,"score_detail":203,"sources":207,"tags":209,"search_phrases":212,"slug":215,"view_count":36,"doi":9,"paper":216,"created_at":225},2610,"整合人工智能、物联网与遥感技术的大田作物智能灌溉管理 综述","https:\u002F\u002Fwww.ebiotrade.com\u002Fnewsf\u002F2026-9\u002F20260913082658847.htm","发表于Biosystems Engineering。对人工智能(AI)、物联网(IoT)和遥感(RS)技术在灌溉管理中的应用进行全面且结构化分析，特别是在优化基于天气、土壤和作物的灌溉调度方面。智能灌溉系统实现了水资源节约(用水量减少高达20-60%)、降低能源消耗和提高作物生产力。未来研究应优先考虑成本效益高的传感器开发和用户友好的AI界面。","Biosystems Engineering","2026-09-13T01:00:00Z",83,{"impact":75,"substance":204,"depth":17,"authority":76,"freshness":205,"relevant":22,"comment":206},21,8,"核心期刊综述，系统梳理AI、物联网与遥感在大田灌溉调度中的融合应用，给出节水20-60%等量化结论，对智慧农业技术路线有参考价值。",[208],{"name":200,"url":198},[27,28,210,211,30],"农业物联网","智能灌溉",[213,214],"农业人工智能 农业物联网 智慧农业 智能灌溉","农业人工智能 农业物联网","农业人工智能农业物联网智慧农业智能灌溉-2610",{"doi":9,"openalex_id":9,"authors":217,"venue":9,"cited_by_count":36,"oa_url":9,"card":218,"direction":222,"ingested_from":224},[],{"tldr":219,"method":220,"finding":221,"direction":222,"opportunity":223},"综述AI、物联网与遥感在大田作物智能灌溉调度中的应用与成效。","结构化综述AI、IoT、RS在基于天气、土壤和作物的灌溉调度中的应用。","智能灌溉可节水20-60%，降低能耗并提高作物生产力。","智慧农业 \u002F 农业物联网","低成本传感器与用户友好AI界面是落地瓶颈，可研究轻量化模型与低成本感知方案。","agent","2026-09-16T00:03:52.381950Z",{"id":227,"title":228,"url":229,"summary":230,"summary_zh":9,"content":9,"source_name":231,"source_url":9,"published_at":201,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":232,"sources":235,"tags":237,"search_phrases":240,"slug":242,"view_count":36,"doi":9,"paper":243,"created_at":250},2609,"AI Framework Estimates Crop Water Stress from Satellite Data in Egypt's Nile Delta","https:\u002F\u002Ffirat.rw\u002Farticles\u002Fai-framework-estimates-crop-water-stress-from-satellite-data-in-egypts-nile-delta","Elbeltagi等发表在Smart Agricultural Technology。研究使用MODIS卫星产品(2018-2025)构建最佳子集回归(BSR)与机器学习相结合的特征优化人工智能框架，估计埃及Dakahliyah省农业植被水分胁迫。NDVI成为EVI的压倒性主导预测因子，相关系数0.934。Random Forest在2024-2025测试期达到0.9943的相关系数，平均绝对误差仅0.0063，根均方误差0.0161，相对绝对误差降至5%以下。","Smart Agricultural Technology",{"impact":233,"substance":75,"depth":17,"authority":20,"freshness":205,"relevant":22,"comment":234},16,"基于MODIS长时序数据与机器学习融合的水分胁迫估算框架，方法新颖、精度高，对干旱区精准灌溉有参考价值，但属区域性案例研究，影响范围有限。",[236],{"name":231,"url":229},[27,28,30,238,239],"水分胁迫","灌溉管理",[241,107],"农业人工智能 智慧农业 水分胁迫 灌溉管理","农业人工智能智慧农业水分胁迫灌溉管理-2609",{"doi":9,"openalex_id":9,"authors":244,"venue":9,"cited_by_count":36,"oa_url":9,"card":245,"direction":57,"ingested_from":224},[],{"tldr":246,"method":247,"finding":248,"direction":57,"opportunity":249},"用MODIS数据与机器学习框架估算埃及尼罗河三角洲农田水分胁迫。","MODIS时序数据，最佳子集回归结合随机森林做特征优化。","NDVI是主导预测因子，随机森林测试相关系数达0.9943，误差低于5%。","可迁移至其他干旱区验证特征优选框架的普适性，并融合多源遥感提升胁迫早期预警。","2026-09-16T00:03:52.319524Z",{"id":252,"title":253,"url":254,"summary":255,"summary_zh":256,"content":9,"source_name":257,"source_url":254,"published_at":258,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":259,"sources":261,"tags":263,"search_phrases":266,"slug":269,"view_count":36,"doi":270,"paper":271,"created_at":287},2541,"Evidence integrity and review utility in crop-image decision support: an audit-to-review framework for precision agriculture","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs42269-026-01492-x","Abstract Background Agricultural artificial intelligence benchmarks can overstate decision reliability when multiple files represent the same biological evidence or when split provenance is unclear. We evaluated a classifier-agnostic audit-to-review framework that fixes evidence identity and role provenance before model comparison, declares aggregation estimands, and links predictive uncertainty to review workload. A public wheat-leaf corpus of 7,595 files was hashed to 5,118 unique contents; eligibility criteria defined a 915-content five-class task. Archived file-level benchmarks were separated from three frozen near-duplicate-grouped P2 partition configurations, and semantic, texture and fusion representations were evaluated under partition-weighted and one-content-one-weight estimands. Split conformal prediction was assessed by empirical coverage, review rate, error capture, review yield and dimensionless review utility. Results Macro-F1 was 0.962 under the legacy P0 benchmark, 0.805 ± 0.030 for semantic-only P2 and 0.819 ± 0.029 for direct fusion. The descriptive P0-P2 workflow-sensitivity gap was approximately 15.7% points, versus a 1.4-point partition-weighted fusion increment (95% bootstrap interval − 0.006 to 0.036). The leading fusion-related variants were numerically close, so no inferential ranking is claimed. At α = 0.10, global split conformal prediction achieved 0.896 empirical coverage while reviewing 20.8% of images and capturing 51.4% of top-1 errors; class-conditional Mondrian calibration achieved 0.923 coverage while reviewing 29.7% and capturing 64.8% of errors. Conclusions Defining the evidence unit is part of the statistical specification of a crop-image benchmark, not merely a data-cleaning step. In this corpus, benchmark interpretation was much more sensitive to evidence-workflow specification than to the evaluated representation refinement. The framework supports auditable internal evaluation and review triage but does not establish field, farm-level or intervention performance.","摘要 背景 当多个文件代表同一生物学证据或划分来源不明时，农业人工智能基准可能高估决策可靠性。我们评估了一个与分类器无关的“审计到复核”框架，该框架在模型比较前固定证据身份和角色来源，声明聚合估计目标，并将预测不确定性关联到复核工作量。一个包含7,595个文件的公开小麦叶片语料库经哈希处理得到5,118个唯一内容；资格标准定义了一个915个内容的五分类任务。将归档的文件级基准与三种冻结的近重复分组P2划分配置分离，并在划分加权和“一内容一权重”估计目标下评估了语义、纹理和融合表示。通过经验覆盖率、复核率、错误捕获率、复核产出率和无量纲复核效用评估分裂保形预测。结果 在遗留P0基准下，宏F1为0.962；仅语义P2为0.805 ± 0.030；直接融合为0.819 ± 0.029。描述性的P0-P2工作流敏感性差距约为15.7个百分点，而划分加权融合增量为1.4个百分点（95%自助法区间−0.006至0.036）。领先的融合相关变体在数值上接近，因此不声称推断性排序。在α = 0.10时，全局分裂保形预测实现了0.896的经验覆盖率，同时复核了20.8%的图像并捕获了51.4%的top-1错误；类条件Mondrian校准实现了0.923的覆盖率，同时复核了29.7%并捕获了64.8%的错误。结论 定义证据单元是作物图像基准统计规范的一部分，而不仅仅是数据清理步骤。在该语料库中，基准解释对证据-工作流规范的敏感性远高于对所评估表示改进的敏感性。该框架支持可审计的内部评估和复核分诊，但不能确立田间、农场级或干预性能。","Bulletin of the National Research Centre\u002FBulletin of the National Research Center","2026-09-14T00:00:00Z",{"impact":152,"substance":75,"depth":17,"authority":20,"freshness":21,"relevant":22,"comment":260},"该论文提出面向作物图像决策支持的证据完整性审计框架，揭示基准测试中证据单元定义对性能评估的显著影响，方法新颖、数据规模较大，对农业AI模型评估具有参考价值，但属细分领域方法学研究，产业影响有限。",[262],{"name":257,"url":254},[27,28,30,264,265],"作物病害识别","模型评估",[267,268],"作物病害识别 农业人工智能 智慧农业 模型评估","作物病害识别 农业人工智能","作物病害识别农业人工智能智慧农业模型评估-2541","10.1186\u002Fs42269-026-01492-x",{"doi":270,"openalex_id":272,"authors":273,"venue":257,"cited_by_count":36,"oa_url":254,"card":282,"direction":138,"ingested_from":59},"W7212800882",[274,277,279],{"name":275,"orcid":276},"Xin Li","https:\u002F\u002Forcid.org\u002F0009-0005-0670-5006",{"name":278,"orcid":9},"Bojian Guo",{"name":280,"orcid":281},"A.Dzh. Kartanova","https:\u002F\u002Forcid.org\u002F0000-0003-1479-0747",{"tldr":283,"method":284,"finding":285,"direction":138,"opportunity":286},"提出审计到评审框架，先固定证据身份再比较作物图像分类器，并关联预测不确定性与评审工作量。","对7595个小麦叶片文件哈希去重，构建915内容五类任务，用分裂保形预测评估。","基准解释对证据工作流规范远比对表示优化敏感，P0与P2宏F1差约15.7个百分点。","可将证据单元定义与评审效用纳入农业AI基准标准，并探索田间级验证与主动学习结合。","2026-09-15T23:30:34.690905Z"]