[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1824":3,"related-1824":97},{"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":30,"secondary_topics":36,"classification":37,"search_phrases":62,"slug":65,"view_count":21,"doi":66,"paper":67,"created_at":96},1824,"Multi-attribute spectral imaging with a wavelength-aware band-gated vision transformer for camellia oil adulteration detection","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1931619","Adulteration-induced differences in near-infrared spectra of camellia oil may occur simultaneously in the global spectral profile, local variation patterns, and interactions among distant spectral regions. Conventional one-dimensional modeling provides limited means to preserve and jointly learn these multilevel relationships. This study proposes a wavelength-aware Vision Transformer driven by multi-attribute spectral imaging (ViT-BandGate) for qualitative detection of camellia oil adulteration. The preprocessed spectrum and its first- and second-order variations are encoded into a two-dimensional joint representation that preserves their wavelength-wise co-location, thereby integrating global absorption profiles with local slope and curvature information. A Transformer subsequently learns long-range dependencies across spectral regions, while the proposed BandGate adaptively modulates regional token contributions under an explicit wavelength-continuity prior. The method was evaluated on a multi-date dataset containing 884 near-infrared spectra: 84 pure camellia-oil spectra and 800 spectra adulterated with rapeseed, peanut, or sunflower oil. Using a stratified hold-out test set and a validation-calibrated decision threshold, ViT-BandGate achieved 98.29% accuracy, 99.07% balanced accuracy, 100.00% pure-oil recall, 98.14% adulterated-oil recall, and 99.91% ROC AUC. The results demonstrate high sensitivity to adulterated oils while maintaining reliable recognition of pure camellia-oil samples. Compared with direct one-dimensional deep models and conventional baselines, ViT-BandGate achieved the strongest balanced screening profile for qualitative adulteration detection. These findings indicate that preserving wavelength-wise co-location among multi-order spectral attributes and jointly learning regional importance and global dependencies can improve qualitative adulteration recognition. The proposed framework provides a structured spectral-representation strategy for qualitative, reagent-free authenticity screening of edible oils.","掺假引起的山茶油近红外光谱差异可能同时体现在全局光谱轮廓、局部变化模式以及远距离光谱区域间的交互作用中。传统一维建模方法在保留并联合学习这些多层次关系方面能力有限。本研究提出一种由多属性光谱成像驱动的波长感知视觉Transformer（ViT-BandGate），用于山茶油掺假的定性检测。将预处理后的光谱及其一阶和二阶变化编码为二维联合表示，该表示保留了波长维度的共定位关系，从而整合了全局吸收轮廓与局部斜率和曲率信息。随后，Transformer学习跨光谱区域的长程依赖关系，而所提出的BandGate模块在显式波长连续性先验下自适应调制区域令牌的贡献。该方法在一个包含884条近红外光谱的多日期数据集上进行了评估：其中84条为纯山茶油光谱，800条为掺入菜籽油、花生油或葵花籽油的掺假光谱。使用分层留出测试集和验证校准的决策阈值，ViT-BandGate实现了98.29%的准确率、99.07%的平衡准确率、100.00%的纯油召回率、98.14%的掺假油召回率以及99.91%的ROC AUC。结果表明，该方法对掺假油具有高灵敏度，同时保持了对纯山茶油样品的可靠识别。与直接的一维深度模型和传统基线方法相比，ViT-BandGate在定性掺假检测中展现出最强的平衡筛查性能。这些发现表明，保留多阶光谱属性间的波长共定位关系，并联合学习区域重要性与全局依赖关系，能够提升定性掺假识别的效果。所提出的框架为食用油的无试剂定性真实性筛查提供了一种结构化的光谱表示策略。",null,"Frontiers in Sustainable Food Systems","2026-09-07T00:00:00Z","论文",10,false,78,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,12,8,1,"提出ViT-BandGate模型用于山茶油掺假检测，方法新颖，数据可靠，对食品安全有实际价值。",[24],{"name":10,"url":6},[26,27,28,29],"智慧农业","农业人工智能","食用油检测","光谱成像",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},"A4","soybean-oilseeds","大豆与油料","\u002Ftopic\u002Ffield-crops\u002Fsoybean-oilseeds","A",[],{"status":38,"locked":14,"version":39,"method":40,"evidence":41},"classified","2026-10-22.2","rules",{"candidates":42,"agricultural_evidence":45},[43,52,57],{"code":31,"score":44,"supported":45,"title":46,"abstract":47,"tags":51},9,true,[],[48,49,50],"花生","rapeseed","peanut",[],{"code":53,"score":21,"supported":14,"title":54,"abstract":55,"tags":56},"I1",[],[],[26],{"code":58,"score":21,"supported":14,"title":59,"abstract":60,"tags":61},"I2",[],[],[27],[63,64],"农业人工智能 食用油检测 光谱成像 智慧农业","农业人工智能 食用油检测","农业人工智能食用油检测光谱成像智慧农业-1824","10.3389\u002Ffsufs.2026.1931619",{"doi":66,"openalex_id":68,"authors":69,"venue":10,"cited_by_count":88,"oa_url":6,"card":89,"direction":93,"ingested_from":95},"W7210269982",[70,72,74,76,78,80,83,85],{"name":71,"orcid":9},"刘晓政",{"name":73,"orcid":9},"Yingxue Ma",{"name":75,"orcid":9},"Haoxia Huang",{"name":77,"orcid":9},"Xiaohui Wan",{"name":79,"orcid":9},"Jiahao Zhu",{"name":81,"orcid":82},"Lijie Chen","https:\u002F\u002Forcid.org\u002F0000-0003-4457-3727",{"name":84,"orcid":9},"Peng Zheng",{"name":86,"orcid":87},"Shangdong Zheng","https:\u002F\u002Forcid.org\u002F0000-0002-7444-3459",0,{"tldr":90,"method":91,"finding":92,"direction":93,"opportunity":94},"提出波长感知的ViT-BandGate模型，利用多属性光谱成像检测山茶油掺假，实现高精度识别。","将光谱及一阶、二阶导数编码为二维表示，用Transformer学习长程依赖，Ba","ViT-BandGate在884个近红外光谱上达98.29%准确率，纯油召回100%，优于一维模型。","农业人工智能与决策模型","可探索将多属性光谱成像与Transformer结合用于其他食用油或农产品的掺假检测，并优化模型轻量化。","openalex","2026-09-07T23:30:07.266754Z",{"total":98,"page":21,"page_size":98,"items":99},6,[100,180,238,337,402,458],{"id":101,"title":102,"url":103,"summary":104,"summary_zh":105,"content":9,"source_name":106,"source_url":103,"published_at":107,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":108,"score_detail":109,"sources":114,"tags":116,"primary_topic":120,"secondary_topics":121,"classification":127,"search_phrases":146,"slug":149,"view_count":142,"doi":150,"paper":151,"created_at":179},3986,"YOLOv8-MST: a lightweight and efficient real-time detection and counting deep learning algorithm for double-grain peanut seeding","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112481","YOLOv8-MST: a lightweight and efficient real-time detection and counting deep learning algorithm for double-grain peanut seeding。Computers and Electronics in Agriculture","YOLOv8-MST：一种用于双粒花生播种的轻量高效实时检测与计数深度学习算法。《计算机与电子农业》","Computers and Electronics in Agriculture","2026-10-01T00:00:00Z",72,{"impact":19,"substance":110,"depth":111,"authority":112,"freshness":44,"relevant":21,"comment":113},20,17,14,"核心期刊发表的轻量化实时检测计数算法，对花生精量播种智能化有实用价值，但属细分技术进展，影响面有限。",[115],{"name":106,"url":103},[26,27,117,118,119],"目标检测","精准农业","花生播种",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[122],{"code":58,"slug":123,"name":124,"path":125,"parent_code":126},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":38,"locked":14,"version":39,"method":40,"evidence":128},{"candidates":129,"agricultural_evidence":45},[130,137,141],{"code":58,"score":131,"supported":45,"title":132,"abstract":134,"tags":136},16,[133],"deep learning",[135,133],"深度学习",[27,117],{"code":31,"score":112,"supported":45,"title":138,"abstract":139,"tags":140},[50],[48,50],[],{"code":53,"score":142,"supported":14,"title":143,"abstract":144,"tags":145},2,[],[],[26,118],[147,148],"YOLOv8-MST 花生播种","双粒花生 实时检测计数","YOLOv8-MST花生播种-3986","10.1016\u002Fj.compag.2026.112481",{"doi":150,"openalex_id":152,"authors":153,"venue":106,"cited_by_count":88,"oa_url":9,"card":174,"direction":93,"ingested_from":95},"W7215133023",[154,157,159,161,163,165,167,169,172],{"name":155,"orcid":156},"Baogang Li","https:\u002F\u002Forcid.org\u002F0000-0002-3430-7508",{"name":158,"orcid":9},"Enxi Liu",{"name":160,"orcid":9},"Yijia Chen",{"name":162,"orcid":9},"Zhipeng Feng",{"name":164,"orcid":9},"Yuhuan Li",{"name":166,"orcid":9},"Zunmin Liu",{"name":168,"orcid":9},"Jingchun Tang",{"name":170,"orcid":171},"Jiang Liu","https:\u002F\u002Forcid.org\u002F0000-0001-6746-9987",{"name":173,"orcid":9},"Fazhan Yang",{"tldr":175,"method":176,"finding":177,"direction":93,"opportunity":178},"提出YOLOv8-MST轻量算法，实现双粒花生播种实时检测与计数。","基于YOLOv8改进的轻量深度学习模型，用于花生播种图像检测计数。","YOLOv8-MST在保持高精度的同时实现轻量化实时检测与计数。","可探索轻量模型在多种作物播种质量实时监测中的泛化与边缘部署。","2026-10-02T23:30:03.835851Z",{"id":181,"title":182,"url":183,"summary":184,"summary_zh":185,"content":9,"source_name":106,"source_url":183,"published_at":107,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":186,"score_detail":187,"sources":189,"tags":191,"primary_topic":194,"secondary_topics":195,"classification":196,"search_phrases":213,"slug":216,"view_count":217,"doi":218,"paper":219,"created_at":237},3971,"A gradient-enhanced lightweight network for peanut leaf disease segmentation in complex field scenarios","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.compag.2026.112482","A gradient-enhanced lightweight network for peanut leaf disease segmentation in complex field scenarios。Computers and Electronics in Agriculture","一种用于复杂田间场景下花生叶部病害分割的梯度增强轻量级网络",69,{"impact":19,"substance":17,"depth":131,"authority":112,"freshness":44,"relevant":21,"comment":188},"发表于农业信息领域核心期刊的轻量级分割网络研究，方法有创新但属细分技术进展，公共影响有限。",[190],{"name":106,"url":183},[26,27,192,48,193],"病虫害识别","图像分割",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":197},{"candidates":198,"agricultural_evidence":45},[199,204,209],{"code":31,"score":200,"supported":45,"title":201,"abstract":202,"tags":203},15,[50],[48,50],[48],{"code":58,"score":205,"supported":14,"title":206,"abstract":207,"tags":208},3,[],[],[27,193,192],{"code":53,"score":21,"supported":14,"title":210,"abstract":211,"tags":212},[],[],[26],[214,215],"花生叶片病害 图像分割","花生 病害识别 轻量网络","花生叶片病害图像分割-3971",4,"10.1016\u002Fj.compag.2026.112482",{"doi":218,"openalex_id":220,"authors":221,"venue":106,"cited_by_count":88,"oa_url":9,"card":232,"direction":93,"ingested_from":95},"W7215079154",[222,225,228,230],{"name":223,"orcid":224},"Song Liu","https:\u002F\u002Forcid.org\u002F0009-0002-6649-4990",{"name":226,"orcid":227},"Chunjian Hua","https:\u002F\u002Forcid.org\u002F0000-0001-5397-2688",{"name":229,"orcid":9},"俞建峰 Yu Jianfeng",{"name":231,"orcid":9},"Ying Chen",{"tldr":233,"method":234,"finding":235,"direction":93,"opportunity":236},"提出梯度增强轻量网络，用于复杂田间场景下花生叶部病害分割。","梯度增强模块与轻量网络结合，在复杂田间图像上分割花生叶部病害。","该方法在复杂田间场景下实现花生叶部病害的轻量高效分割。","可探索轻量分割模型在移动端部署及多作物病害泛化能力。","2026-10-02T23:30:02.241779Z",{"id":239,"title":240,"url":241,"summary":242,"summary_zh":9,"content":243,"source_name":244,"source_url":9,"published_at":245,"category":246,"cover_url":9,"hotness":13,"is_selected":45,"score":247,"score_detail":248,"sources":254,"tags":256,"primary_topic":261,"secondary_topics":262,"classification":263,"search_phrases":332,"slug":335,"view_count":98,"doi":9,"paper":9,"created_at":336},3924,"山东省人民政府办公厅印发《关于大力发展智慧农业的实施意见》（鲁政办字〔2026〕85号）","https:\u002F\u002Fny.zibo.gov.cn\u002Fart\u002F2026\u002F9\u002F30\u002Fart_5879_3024122.html","《意见》提出到2030年山东将建设80个数字农业发展县，重点打造300个高水平智慧农业应用场景，全省农业生产信息化率达到60%以上，粮油作物水肥一体化应用面积达到2000万亩。围绕\"加力推进优势产业智慧转型\"，《意见》部署智慧赋能主要作物单产提升、设施农业改造升级、畜牧业节本增效、渔业产业提档升级四大方向，加快\"北斗+智慧农业\"应用推广，构建小麦、玉米、大豆、花生等主要作物数字化种植技术体系；建设省级种质资源数据库平台，推动经验育种向智能设计育种转变。","*   [](https:\u002F\u002Fny.zibo.gov.cn\u002Fart\u002F2026\u002F9\u002F30\u002Fart_5879_3024122.html)\n\n*   [](https:\u002F\u002Fny.zibo.gov.cn\u002Fart\u002F2026\u002F9\u002F30\u002Fart_5879_3024122.html)\n\n[首页](https:\u002F\u002Fny.zibo.gov.cn\u002Findex.html)>[三农业务](https:\u002F\u002Fny.zibo.gov.cn\u002Fcol\u002Fcol980\u002Findex.html)>[政策法规](https:\u002F\u002Fny.zibo.gov.cn\u002Fcol\u002Fcol5879\u002Findex.html)\n\n山东省人民政府办公厅关于大力发展智慧农业的实施意见\n* * *\n发布日期： 2026-09-30 13:58:48 浏览次数:字体：[ [大](javascript:doZoom(17))[中](javascript:doZoom(15))[小](javascript:doZoom(13)) ]\n鲁政办字〔2026〕85号\n\n各市人民政府，各县（市、区）人民政府，省政府有关部门、有关直属机构，有关高等院校：\n\n为贯彻落实党中央、国务院关于智慧农业的决策部署，深入实施“人工智能＋农业”行动，加快推进物联网、大数据、人工智能、机器人等现代信息技术在农业农村领域全方位、全链条推广应用，打造全国智慧农业发展高地，助力农业农村现代化，现提出如下实施意见。\n\n**一、加力推进优势产业智慧转型**\n\n（一）智慧赋能主要作物单产提升。利用卫星遥感、无人机、地面传感等技术，统筹布局“四情”监测网络，深化农业气象灾害风险减量工作，提升实时感知和预警能力。推进农业生产数字化改造，大力推广应用智能农业机械化、水肥一体化智慧管控等技术，加强农田新型基础设施建设和改造升级，探索粮食减损智能系统推广，到2030年粮油作物水肥一体化应用面积达到2000万亩。促进良田、良种、良机、良法与数字化技术有机融合，开展“北斗＋智慧农业”支撑粮油作物大面积单产提升应用推广，构建小麦、玉米、大豆、花生等主要作物数字化种植技术体系，提升大面积单产水平。（牵头单位：省农业农村厅，责任单位：省大数据局、省气象局）\n\n（二）智慧赋能设施农业改造升级。有序推进集中连片老旧低效设施蔬菜基地、果园、茶园等更新改造，建设标准化、智慧化现代设施。加快精准感知、智能调控等数字化技术装备集成应用，推广应用设施农业智能体，实现全生长周期智能化决策。引导规模化设施农业经营主体，部署应用生产经营全过程数字化管理平台，推进设施农业智慧化管理。（牵头单位：省农业农村厅，责任单位：省发展改革委、省自然资源厅、省大数据局）\n\n（三）智慧赋能畜牧业节本增效。引导规模养殖场（户）进行数字化改造，建立电子养殖档案，推进养殖、饲料兽药、屠宰等环节数据互联互通，实现直联直报。鼓励规模养殖场（户）、畜牧屠宰企业加强重点环节监测预警，完善从生产到屠宰的全链条数字化监管。大力推广精准饲喂、智能环控、自动巡检、个体识别等智慧化养殖装备与技术，全面提高综合养殖效益。（牵头单位：省畜牧局，责任单位：省农业农村厅、省大数据局）\n\n（四）智慧赋能渔业产业提档升级。分类有序推进池塘、工厂化等养殖模式智能化改造，科学配备饲料精准投喂、智能增氧、循环水处理、水质在线监测等智能化设施。深化海洋牧场现代化建设，因地制宜推广应用水域环境监测、水下巡检、智能采捕等先进技术装备。推进苗种繁育、健康养殖、产品加工、冷链物流、综合服务等全链条数字化融合发展。强化渔政执法数字化建设，加快海洋渔船设施装备更新改造，利用渔船监控平台，推广AIS防避碰和北斗船位终端的应用，保障渔业生产安全。（牵头单位：省农业农村厅，责任单位：省工业和信息化厅）\n\n**二、加力推进关键领域智慧化建设**\n\n（五）大力发展智能农机装备。加快发展适用于精准作业、果蔬采摘、病虫害巡检、畜禽健康监测、深海养殖监控等复杂场景的智能农机、农业机器人等智能装备。扩大无人机在植保、施肥、农情监测等方面普及应用。鼓励农机制造企业在农机装备出厂前加装北斗导航、传感器等智能化设备，推动农业基础设施与智能农机协同联动。支持“无人农场”建设。（牵头单位：省工业和信息化厅，责任单位：省科技厅、省农业农村厅、省畜牧局）\n\n（六）加快智慧育种技术应用。加大种质资源数据收集与利用力度，建设省级种质资源数据库平台。推动人工智能、大数据等新一代信息技术与传统育种深度融合。支持科研机构和种业企业联合打造智能育种平台，开发智能设计育种工具，推动经验育种向智能设计育种转变，加速育种创新。以寿光、兰陵、莘县、济阳等蔬菜生产大县为重点，推动集约化数字种苗工厂建设。（省发展改革委、省科技厅、省农业农村厅按职责分工负责）\n\n（七）推进农业农村管理服务数字化。建好用好“齐鲁农云”山东省数字农业农村综合管理服务平台，支持开展省市共建试点工作，打造覆盖全省、上下贯通、业务协同的农业农村“数字大脑”。拓展“鲁农码”应用领域，实现涉农业务“一码通行”、服务事项“一码通办”，推进全省农业农村用地“一张图”管理。推动现代农事综合服务中心数字化建设，促进农业社会化服务资源高效整合与供需智能匹配，推动涉农数据价值转化。（省自然资源厅、省农业农村厅、省大数据局按职责分工负责）\n\n（八）加快全产业链数字化改造。充分发挥农业“链主”企业、龙头企业等示范作用，带动产业链上下游各类经营主体进行数字化改造。支持农产品精深加工数字化技术及智能装备研发与应用，推动人工智能技术在农产品分选、加工过程质量控制、供应链协同优化等关键环节的深度融合，建设特色农业“大脑”。积极推动农产品质量安全承诺达标合格证信息化，推进“区块链＋”在农产品质量安全管理中的应用，探索建立食用农产品全链条信息化追溯体系。强化农产品市场监测预警，引导产业有序发展。建好用好“齐鲁农超”平台，加快推动农村电商平台经济高质量发展，不断提升“好品山东 优特农品”品牌影响力。（省发展改革委、省工业和信息化厅、省农业农村厅、省商务厅、省市场监管局、省畜牧局按职责分工负责）\n\n**三、加力推进数字生态效能释放**\n\n（九）强化数据治理。健全农业农村数据管理制度，完善数据资源目录，加强涉农数据治理，建设农业农村高质量数据集。探索构建涉农数据资产价值转化路径，推动数据在农业生产、经营、管理、服务等各环节的应用，充分发挥数据要素价值。搭建安全可信的数据共享与交易环境，构建农业行业可信数据空间，推进网络安全与数据安全防护体系建设。（省委网信办、省农业农村厅、省大数据局按职责分工负责）\n\n（十）强化技术创新。实施“人工智能＋农业”行动，鼓励企业与科研院所、高校联合开展农业传感器与专用芯片、模型算法等研发攻关与成果转化，深入推进人工智能大模型、大数据分析等技术在农业农村领域融合应用，加快构建“空天地一体化”数字农业技术体系。鼓励各地根据发展需求，有针对性地组织开展智慧农业技术研发。强化企业创新主体地位，积极培育智慧农业领军企业，整合科研院所、高校、龙头企业和新型生产经营组织等资源，形成“产学研用”一体化创新模式。（省教育厅、省科技厅、省农业农村厅、省畜牧局按职责分工负责）\n\n（十一）强化技术推广。鼓励涉农高校、科研机构、推广单位与经营主体开展智慧农业技术集成应用示范，实施省农业重大技术协同推广计划，遴选发布一批智慧农业主推技术。加快推进物联网、大数据、人工智能、北斗导航、高分遥感等现代信息技术推广应用。加快制（修）订智慧农业标准规范，建立健全智慧农业技术推广服务体系，加强智慧农业技术指导和推广。鼓励有条件的农业社会化服务主体加快应用智慧农业技术，提供农情遥感监测、农事作业、经营管理、防灾减灾等技术服务。（省教育厅、省科技厅、省农业农村厅、省畜牧局按职责分工负责）\n\n（十二）强化场景应用。开展实数融合促进行动，依托全省农业数字赋能平台等，推进数字农业发展县建设，探索形成区域性的智慧农业解决方案。加快粮食作物、设施园艺、畜牧水产、特色产业及智慧气象、农机服务等重点领域的场景创新与推广，指导各地根据产业发展实际，因地制宜打造一批技术融合度高、应用实效性强、复制推广性好的智慧农业应用场景。到2030年，建设80个数字农业发展县，重点打造300个高水平的智慧农业应用场景，全省农业生产信息化率达到60％以上。（省农业农村厅、省大数据局、省畜牧局、省气象局按职责分工负责）\n\n**四、切实做好组织保障工作**\n\n由农业农村部门牵头，加强与发展改革、财政、科技、工业和信息化、自然资源、畜牧兽医、大数据、气象等部门沟通配合，强化省、市、县三级协同联动，确保任务落到实处。统筹乡村振兴重大专项资金等各级相关领域财政资金推进智慧农业建设，重点支持关键技术攻关、数字农业发展县和应用场景建设。省级每年支持建设30个智慧农业应用场景。优化金融服务落实税收优惠政策，支持智慧农业科技企业和创新平台发展。推动高校、科研院所与智慧农业企业深化合作，培养精通信息技术与农业产业的复合型人才。持续推动高素质农民培育、基层农技推广队伍素质提升等项目，提升生产经营主体数字素养与技能。推进智慧农业领域从业人员职业技能培训与评价，吸引各类人才投身智慧农业发展。（省委金融办、省发展改革委、省教育厅、省科技厅、省工业和信息化厅、省财政厅、省人力资源社会保障厅、省自然资源厅、省农业农村厅、省大数据局、省畜牧局、山东金融监管局、省税务局、省气象局按职责分工负责）\n\n山东省人民政府办公厅\n\n2026年9月18日\n[【打印本页】](javascript:window.print())[【关闭窗口】](javascript:window.opener=null;window.open('','_self');window.close();)","淄博市农业农村局·2026年09月30日","2026-09-30T05:58:48Z","政策",90,{"impact":249,"substance":250,"depth":251,"authority":252,"freshness":20,"relevant":21,"comment":253},26,24,19,13,"省级智慧农业顶层政策文件，条款与量化目标明确，对全国农业信息化建设具有示范参考价值，值得进入每日精选。",[255],{"name":244,"url":241},[26,27,257,258,259,260],"智能农机","齐鲁农云","农业数据要素","数字农业发展县",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":264},{"candidates":265,"agricultural_evidence":45},[266,271,276,282,288,294,300,306,312,318,323,327],{"code":53,"score":200,"supported":45,"title":267,"abstract":268,"tags":270},[26],[26,269],"数字农业",[26],{"code":31,"score":98,"supported":45,"title":272,"abstract":273,"tags":275},[],[274,48],"大豆",[],{"code":277,"score":205,"supported":45,"title":278,"abstract":279,"tags":281},"A2",[],[280],"小麦",[],{"code":283,"score":205,"supported":45,"title":284,"abstract":285,"tags":287},"A3",[],[286],"玉米",[],{"code":289,"score":205,"supported":45,"title":290,"abstract":291,"tags":293},"A8",[],[292],"单产提升",[],{"code":295,"score":205,"supported":45,"title":296,"abstract":297,"tags":299},"C",[],[298],"畜牧",[],{"code":301,"score":205,"supported":45,"title":302,"abstract":303,"tags":305},"D",[],[304],"渔业",[],{"code":307,"score":205,"supported":45,"title":308,"abstract":309,"tags":311},"F1",[],[310],"种质资源",[],{"code":313,"score":205,"supported":45,"title":314,"abstract":315,"tags":317},"F3",[],[316],"水肥一体化",[],{"code":319,"score":21,"supported":14,"title":320,"abstract":321,"tags":322},"F5",[],[],[257],{"code":58,"score":21,"supported":14,"title":324,"abstract":325,"tags":326},[],[],[27],{"code":328,"score":21,"supported":14,"title":329,"abstract":330,"tags":331},"I6",[],[],[259],[333,334],"山东 智慧农业 实施意见","鲁政办字 85号 智慧农业","山东智慧农业实施意见-3924","2026-10-02T11:53:08.547636Z",{"id":338,"title":339,"url":340,"summary":341,"summary_zh":9,"content":9,"source_name":342,"source_url":9,"published_at":343,"category":12,"cover_url":9,"hotness":13,"is_selected":45,"score":344,"score_detail":345,"sources":348,"tags":350,"primary_topic":355,"secondary_topics":356,"classification":357,"search_phrases":389,"slug":392,"view_count":217,"doi":9,"paper":393,"created_at":401},3125,"Full-Season Agentic Farm System FAIRY: Event-Driven Multi-Agent Orchestration for Soybean Production（FAIRY 全季节智能体农场系统：大豆生产的事件驱动多智能体编排）","https:\u002F\u002Faiagentstore.ai\u002Fai-agent-news\u002Ftopic\u002Fagriculture-food\u002F2026-08-11","哈尔滨工业大学研究人员发布并部署全栈、事件驱动的智能体引擎 FAIRY 于中国运行中的大豆研究农场。FAIRY 集成传感器、无人机、卫星植被产品、机械 API、作物过程模型与多智能体编排层，执行起垄→播种→灌溉→病虫害防治→收获→干燥→存储工作流，并在 64 垄研究场上跨 100 个全季节场景评估 9 个智能体控制器。这是智能体系统能够在大农业时间尺度和延迟结果下进行推理的最清晰演示之一，将农业中的智能体工作从实验室演示推进到全过程评估。同期 arXiv 推出 HarvestBench 基准将 LLM 驱动智能体置于农场网格世界（拖拉机面临动物选择绕行或碾压），结果显示模型差异巨大、对道德简报高度敏感、避免意愿具有可衡量的价格弹性。","Harbin Institute of Technology \u002F arXiv","2026-09-18T00:00:00Z",89,{"impact":250,"substance":346,"depth":251,"authority":112,"freshness":44,"relevant":21,"comment":347},23,"哈工大在真实大豆农场部署全季节事件驱动多智能体系统并配套 HarvestBench 基准，是农业智能体从演示走向全过程评估的标志性进展，专业深度与信息增量俱佳。",[349],{"name":342,"url":340},[26,351,27,352,353,354],"无人农场","农业遥感","多智能体","大豆生产",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":358},{"candidates":359,"agricultural_evidence":45},[360,365,371,375,380,384],{"code":31,"score":251,"supported":45,"title":361,"abstract":363,"tags":364},[274,362],"soybean",[274],[],{"code":366,"score":205,"supported":45,"title":367,"abstract":368,"tags":370},"H3",[],[369],"灌溉",[],{"code":58,"score":142,"supported":14,"title":372,"abstract":373,"tags":374},[],[],[27,353],{"code":376,"score":21,"supported":14,"title":377,"abstract":378,"tags":379},"F6",[],[],[351],{"code":53,"score":21,"supported":14,"title":381,"abstract":382,"tags":383},[],[],[26],{"code":385,"score":21,"supported":14,"title":386,"abstract":387,"tags":388},"I3",[],[],[352],[390,391],"哈工大 FAIRY 大豆","HarvestBench 智能体 农场","哈工大FAIRY大豆-3125",{"doi":9,"openalex_id":9,"authors":394,"venue":9,"cited_by_count":88,"oa_url":9,"card":395,"direction":93,"ingested_from":400},[],{"tldr":396,"method":397,"finding":398,"direction":93,"opportunity":399},"部署全季节事件驱动多智能体系统FAIRY，在大豆农场完成从起垄到存储的全流程评估。","集成传感器、无人机、卫星、作物模型与机械API，用9个智能体控制器在64垄100","智能体系统能在大农业时间尺度下推理，模型差异大且对道德简报敏感。","可研究多智能体在长周期、延迟反馈农业任务中的鲁棒性与伦理约束机制。","agent","2026-09-22T00:05:38.611717Z",{"id":403,"title":404,"url":405,"summary":406,"summary_zh":9,"content":9,"source_name":407,"source_url":9,"published_at":408,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":409,"sources":411,"tags":413,"primary_topic":417,"secondary_topics":418,"classification":419,"search_phrases":445,"slug":448,"view_count":142,"doi":9,"paper":449,"created_at":457},3000,"Plant-GeoAT：几何感知3D植物点云器官身份解析用于器官级表型分析","https:\u002F\u002Fwww.mdpi.com\u002F2073-4395\u002F16\u002F18\u002F1809","四川农业大学吴俊杰等提出Plant-GeoAT几何感知解析网络，在RGB融合前编码局部相对-XYZ邻域并将空间邻域与特征空间关系耦合用于密集点预测。在自建油菜数据集、基于图像的大豆数据集、激光扫描Pheno4D玉米和番茄数据集上评估，5个随机种子下mIoU分别达92.26±0.28%、82.50±0.24%、99.74±0.05%、94.75±0.15%，玉米Stem IoU达99.57±0.09%。","MDPI Agronomy 16(18):1809","2026-09-15T00:00:00Z",{"impact":17,"substance":346,"depth":17,"authority":252,"freshness":98,"relevant":21,"comment":410},"方法新颖、多作物多数据集验证且精度数据扎实，属器官级表型分析细分领域的重要技术进展，值得进入每日精选。",[412],{"name":407,"url":405},[26,27,414,415,416],"油菜","高通量表型","三维点云",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":420},{"candidates":421,"agricultural_evidence":45},[422,427,431,437,441],{"code":31,"score":423,"supported":45,"title":424,"abstract":425,"tags":426},7,[],[274,414],[414],{"code":283,"score":205,"supported":45,"title":428,"abstract":429,"tags":430},[],[286],[],{"code":432,"score":205,"supported":45,"title":433,"abstract":434,"tags":436},"B1",[],[435],"番茄",[],{"code":58,"score":142,"supported":14,"title":438,"abstract":439,"tags":440},[],[],[416,27],{"code":53,"score":21,"supported":14,"title":442,"abstract":443,"tags":444},[],[],[26],[446,447],"四川农业大学 植物点云 器官识别","Plant-GeoAT 表型分析","四川农业大学植物点云器官识别-3000",{"doi":9,"openalex_id":9,"authors":450,"venue":9,"cited_by_count":88,"oa_url":9,"card":451,"direction":455,"ingested_from":400},[],{"tldr":452,"method":453,"finding":454,"direction":455,"opportunity":456},"提出几何感知网络Plant-GeoAT，实现3D植物点云器官身份解析与器官级表型分析。","编码局部相对XYZ邻域并耦合空间与特征空间关系，在油菜、大豆、玉米、番茄点云数据","五个数据集mIoU最高达99.74%，玉米茎IoU达99.57%，验证了几何感知对器官分割的有效性。","农业遥感与作物表型","可探索跨物种、跨传感器的轻量化几何感知模型，并推动器官级表型与基因型关联分析。","2026-09-20T00:03:08.094512Z",{"id":459,"title":460,"url":461,"summary":462,"summary_zh":9,"content":9,"source_name":463,"source_url":9,"published_at":464,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":465,"sources":467,"tags":469,"primary_topic":473,"secondary_topics":474,"classification":475,"search_phrases":498,"slug":501,"view_count":205,"doi":9,"paper":502,"created_at":509},2902,"TSAFI-DT:可持续性感知花生产量预测数字孪生框架,登MDPI AI 7(9)364","https:\u002F\u002Fwww.mdpi.com\u002F2673-2688\u002F7\u002F9\u002F364","本研究提出TSAFI-DT可回顾验证的、数据驱动的Digital Twin原型,集成时空数据重建、可持续状态表征、分层产量预测、反事实分析与情景模拟。基于1997-2023年印度地区级花生数据,采用贝叶斯优化的XGBoost模型进行一步前瞻产量预测,RMSE=0.171 t\u002Fha,显著优于基线;结合固定效应与合成控制分析,eRAI扩展再生农业指数整合作物多样性、生产力稳定性、土地利用效率和产量趋势,预测产量在可持续性扰动下可提升12.4%。","MDPI AI","2026-09-14T00:00:00Z",{"impact":17,"substance":18,"depth":251,"authority":252,"freshness":98,"relevant":21,"comment":466},"方法新颖、数据规模扎实的农业数字孪生研究，对智慧农业与产量预测领域有参考价值，但属细分学术进展，非产业级事件。",[468],{"name":463,"url":461},[26,27,470,471,472],"数字孪生","可持续农业","花生产量预测",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":476},{"candidates":477,"agricultural_evidence":45},[478,483,488,494],{"code":53,"score":252,"supported":45,"title":479,"abstract":480,"tags":482},[470],[481],"digital twin",[470,26],{"code":31,"score":484,"supported":45,"title":485,"abstract":486,"tags":487},11,[48],[48],[],{"code":489,"score":217,"supported":45,"title":490,"abstract":491,"tags":493},"H8",[],[492],"再生农业",[471],{"code":58,"score":21,"supported":14,"title":495,"abstract":496,"tags":497},[],[],[27],[499,500],"TSAFI-DT 花生 数字孪生","印度 花生 产量预测","TSAFI-DT花生数字孪生-2902",{"doi":9,"openalex_id":9,"authors":503,"venue":9,"cited_by_count":88,"oa_url":9,"card":504,"direction":93,"ingested_from":400},[],{"tldr":505,"method":506,"finding":507,"direction":93,"opportunity":508},"提出可持续性感知数字孪生框架TSAFI-DT，用于印度花生产量预测与情景模拟。","基于1997-2023年印度地区级数据，用贝叶斯优化XGBoost和合成控制分析","XGBoost预测RMSE为0.171 t\u002Fha，可持续性扰动下产量可提升12.4%。","可探索将数字孪生与实时物联网数据结合，实现动态可持续性评估与决策支持。","2026-09-19T00:06:08.754978Z"]