[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2084":3,"related-2084":75},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"primary_topic":32,"secondary_topics":38,"classification":39,"search_phrases":70,"slug":73,"view_count":59,"doi":8,"paper":8,"created_at":74},2084,"中国农科院资划所基于可解释AI构建农产品产地溯源模型 破解光谱判断产地的\"黑箱\"难题","https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F01c47c41992241608017a544cbad3538.htm","中国农科院农业资源与农业区划研究所智慧农业创新团队基于可解释人工智能构建农产品产地溯源模型,以中国六个地区的猕猴桃为对象,融合光谱成像与多模型组合策略,并运用可解释性方法对最优组合模型进行分层和整体解读,在提升溯源准确率的同时揭示光谱数据判断产地的内在逻辑,为光谱技术在农业生产与品质监管中应用提供系统解决思路,成果发表于《农业人工智能》(Artificial Intelligence in Agriculture)。",null,"[English](https:\u002F\u002Fwww.caas.cn\u002Fen)[邮箱](https:\u002F\u002Fmail.caas.cn\u002F)[数字农科院](https:\u002F\u002Fi.caas.cn\u002F)[](https:\u002F\u002Fwww.caas.cn\u002Fcms\u002Fweb\u002Fsearch\u002Findex.jsp?siteID=cdb01dceb46e48488945d4465e90f221&aba=)\n\n官方微信 \n\n![Image 2](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002F020ac4d3d8b6451782a8702c72604336.jpg)农科专家在线微信公众号\n\n![Image 3](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002Fbf6579e4651e437db31a22cd2249b319.jpg)中国农科院微信公众号\n\n[![Image 4](https:\u002F\u002Fwww.caas.cn\u002Fstatic2022\u002Fimages\u002Fmenu_logo.png)](https:\u002F\u002Fwww.caas.cn\u002Findex.htm)\n\n*   [组织机构](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Findex.htm)[走进中国农科院](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fzjnky\u002Findex.htm)[寄语](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fyczc\u002Findex.htm)[院章程](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fzgnykxyzc\u002Findex.htm)[院领导](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fyld\u002Findex.htm)[历任领导](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Flrld\u002Findex.htm)[组织架构](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fzzjg\u002Findex.htm)[院机关](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fyjg\u002Findex.htm)[院属单位](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fysdw\u002Findex.htm)[区域创新中心](https:\u002F\u002Fwww.caas.cn\u002Fzzjg\u002Fqycxzx\u002Findex.htm) \n*   [新闻中心](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Findex.htm)[图片新闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Ftpxw\u002Findex.htm)[时政要闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002Findex.htm)[农科要闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fnkyw\u002Findex.htm)[科研活动](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002Findex.htm)[三农一线](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fhzdznew\u002Findex.htm)[媒体报道](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Findex.htm)[视频农科](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fspnk\u002Findex.htm)[政务新媒体矩阵](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fzwxmtjz\u002Findex.htm) \n*   [科学研究](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Findex.htm)[重大成果](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fzdcgzs\u002Findex.htm)[科研进展](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fkyjz\u002Findex.htm)[科技创新](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fkjcx\u002Findex.htm)[科技奖励](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fkjjl\u002Findex.htm)[农业高端智库](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fnygdzk\u002Findex.htm)[中国农科院重大科学发现](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fzgnkyzdkxfx\u002Findex.htm)[中国农科院重大产品创制、重大技术突破和重大智库报告](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fjstpzkbg\u002Findex.htm)[中国农科院重大科技任务布局](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fzgnkyzdkjrwbj\u002Findex.htm)[中国农科院十大科研进展](https:\u002F\u002Fwww.caas.cn\u002Fkxyj\u002Fzgnkysdkyjz\u002Findex.htm) \n*   [科技支撑](https:\u002F\u002Fwww.caas.cn\u002Fkjzc\u002Findex.htm)[主推成果](https:\u002F\u002Fwww.caas.cn\u002Fkjzc\u002Fcgzhfw\u002Findex.htm)[产业专家团](https:\u002F\u002Fwww.caas.cn\u002Fkjzc\u002Fcyzjt\u002Findex.htm) \n*   [人才教育](https:\u002F\u002Fwww.caas.cn\u002Frcjy\u002Findex.htm)[人才队伍概况](https:\u002F\u002Fwww.caas.cn\u002Frcjy\u002Frcgk\u002Findex.htm)[人才工作概述](https:\u002F\u002Fwww.caas.cn\u002Frcjy\u002Frcgzgs\u002Findex.htm)[专家学者](https:\u002F\u002Fwww.caas.cn\u002Frcjy\u002Fzjxz\u002Findex.htm)[人才动态](https:\u002F\u002Fwww.caas.cn\u002Frcjy\u002Frcdtnew\u002Findex.htm)[研究生教育](https:\u002F\u002Fgs.caas.cn\u002F)[中国农科院人才招聘网](https:\u002F\u002Fzhaopin.caas.cn\u002F#\u002F) \n*   [国际合作](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Findex.htm)[总体概况](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fgjhzgk\u002Findex.htm)[合作伙伴](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fhzhb\u002Findex.htm)[合作平台](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fhzpt\u002Findex.htm)[合作机制](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fhzzz\u002Findex.htm)[国际农业科学计划](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fgjnykxjh\u002Findex.htm)[中国农科院年报](https:\u002F\u002Fwww.caas.cn\u002Fgjhz\u002Fzgnkynb\u002Findex.htm) \n*   [平台基地](https:\u002F\u002Fwww.caas.cn\u002Fptjd\u002Findex.htm)[资源概况](https:\u002F\u002Fwww.caas.cn\u002Fptjd\u002Fzygk\u002Findex.htm)[平台](https:\u002F\u002Fwww.caas.cn\u002Fptjd\u002Fptq\u002Findex.htm)[基地](https:\u002F\u002Fwww.caas.cn\u002Fptjd\u002Fjd\u002Findex.htm) \n*   [信息服务](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Findex.htm)[资源服务](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fzyfw\u002Findex.htm)[数字农科院](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fsznky\u002Findex.htm)[农业科研信息化典型案例](https:\u002F\u002Fwww.caas.cn\u002Fzt\u002Fxxhbp_0\u002Fnykyxxhdxa_0\u002Findex.htm) \n*   [党建文化](https:\u002F\u002Fdw.caas.cn\u002F)\n\n![Image 5](https:\u002F\u002Fwww.caas.cn\u002Fstatic2022\u002Fimages\u002Flogo_s03.png)[](https:\u002F\u002Fwww.caas.cn\u002Findex.htm)[](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Findex.htm)\n\n[返回门户首页](https:\u002F\u002Fwww.caas.cn\u002Findex.htm)[数字农科院](https:\u002F\u002Fi.caas.cn\u002F)[邮箱](https:\u002F\u002Fmail.caas.cn\u002F)\n\n官方微信 \n\n![Image 6](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002F020ac4d3d8b6451782a8702c72604336.jpg)农科专家在线微信公众号\n\n![Image 7](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002Fbf6579e4651e437db31a22cd2249b319.jpg)中国农科院微信公众号\n\n[English](https:\u002F\u002Fwww.caas.cn\u002Fen)[](https:\u002F\u002Fwww.caas.cn\u002Fcms\u002Fweb\u002Fsearch\u002Findex.jsp?siteID=cdb01dceb46e48488945d4465e90f221&aba=)\n\n*    [院网首页](https:\u002F\u002Fwww.caas.cn\u002Findex.htm) \n*    [图片新闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Ftpxw\u002Findex.htm) \n*    [时政要闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002Findex.htm) \n*    [农科要闻](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fnkyw\u002Findex.htm) \n*    [科研活动](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002Findex.htm) \n*    [三农一线](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fhzdznew\u002Findex.htm) \n*    [媒体报道](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Findex.htm) \n*    [视频农科](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fspnk\u002Findex.htm) \n*    [政务新媒体矩阵](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fzwxmtjz\u002Findex.htm) \n\n[返回首页](https:\u002F\u002Fwww.caas.cn\u002Findex.htm)[English](https:\u002F\u002Fwww.caas.cn\u002Fen\u002F)\n\n[首页](https:\u002F\u002Fwww.caas.cn\u002Findex.htm)-[新闻中心](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Findex.htm)-[科研活动](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002Findex.htm)\n\n分享到\n\n### 可解释模型实现农产品精准溯源\n\n发布时间：2026-09-07 _|_ 来源： 中国农业科学院农业资源与农业区划研究所 _|_ 作者：钱建平\n\n字体[小](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F01c47c41992241608017a544cbad3538.htm)[中](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F01c47c41992241608017a544cbad3538.htm)[大](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F01c47c41992241608017a544cbad3538.htm)\n\n近日，中国农业科学院[农业资源与农业区划研究所](http:\u002F\u002Fiarrp.caas.cn\u002F)智慧农业创新团队根据可解释人工智能构建农产品产地溯源模型，在提升溯源准确率的同时揭示了光谱数据判断产地的内在逻辑，相关研究成果发表在《农业人工智能（Artificial Intelligence in Agriculture）》上。\n\n![Image 8: 019.jpg](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2026-09\u002Ffa69a734a4bc4551a1ea07fc7f019a58.jpg)\n\n利用光谱数据判断农产品产地时，现有方法多依赖单一算法，往往只追求预测准确度，却忽略了模型决策逻辑的解释，限制了可靠溯源模型的开发和对判断过程的理解。科研团队以中国六个地区的猕猴桃为对象，融合光谱成像技术与多模型组合策略构建产地溯源模型，并运用可解释性分析方法对最优组合模型进行分层和整体解读，在提升溯源准确率的同时，揭示了光谱数据背后判断产地的内在逻辑。\n\n该研究为破解光谱分析中数据分析模型难以解释的难题提供了系统解决思路，在关键波长选择、模型稳定性提升及预测结果可信度增强等方面具有参考价值，有助于推动光谱技术在农业生产与品质监管中的实际应用。\n\n该研究得到国家重点研发计划、中国农业科学院科技创新工程等项目支持。（通讯员 姬悦）\n\n原文链接：https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.aiia.2026.05.010\n\n打印本页\n\n关闭本页\n\n[院网信息发布与管理](https:\u002F\u002Fwww.caas.cn\u002Fywxxfbygl\u002Findex.htm)[最新动态](https:\u002F\u002Fwww.caas.cn\u002Fywxxfbygl\u002Findex.htm)\n\n*   [黑土地侵蚀沟时空演变规律获揭示](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fkyhd\u002F1b6aad7e04244664a16d4fb7847890d7.htm)2026-09-07  \n*   [时代·人民·文明——中国元首外交的思想启迪](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002F1c34df72522047cda6b6ee907c331b85.htm)2026-09-06  \n*   [国际社会高度评价习近平主席出席2026年上海合作组织峰会](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002Ff40e8e7090de4b21ad820fe48d76e0fa.htm)2026-09-06  \n*   [[科技日报]第五届中国（伊川）谷子产业创新发展暨小米品鉴大会举办](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Fc40190199f2c41408362a42158eca346.htm)2026-09-06  \n*   [[科技日报]植物免组织培养基因编辑技术实施体系构建成功](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Fac681f80be354336ac3a74cad00b53a5.htm)2026-09-05  \n*   [习近平总书记的贺信激励广大少儿文艺工作者耕耘不辍奋发有为](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002Fb5df37a3afb64eff8dac243ad6dfdaf3.htm)2026-09-05  \n*   [[农民日报]新技术引领油菜育种从单点创新走向全域突破](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Fa24294843da4414fb379d49d5a2c8e27.htm)2026-09-05  \n*   [[中国科学报]油料所系统总结植物免组织培养基因编辑技术前沿研究](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002F4bf1622073c74e259891ca35badd5b0c.htm)2026-09-05  \n*   [习近平主席出席2026年上海合作组织峰会并对吉尔吉斯斯坦、埃及进行国事访问纪实](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fszyw\u002F85bd28610ac947a79ccdde21b72e0b29.htm)2026-09-05  \n*   [[中国科学报] AI驱动高效挖掘杀虫蛋白](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fmtxw\u002Fef7c6565a42a41aca8d3a9cbe2eed074.htm)2026-09-05  \n\n*   [](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fspnk\u002Findex.htm)[视频农科](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fspnk\u002Findex.htm) \n*   [](https:\u002F\u002Fwww.caas.cn\u002Ftzgg\u002Findex.htm)[通知公告](https:\u002F\u002Fwww.caas.cn\u002Ftzgg\u002Findex.htm) \n*   [](https:\u002F\u002Fwww.caas.cn\u002Findex.htm?mao=special_attention)[特别关注](https:\u002F\u002Fwww.caas.cn\u002Findex.htm?mao=special_attention) \n*   [](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fzwxmtjz\u002Findex.htm)[政务新媒体矩阵](https:\u002F\u002Fwww.caas.cn\u002Fxwzx\u002Fzwxmtjz\u002Findex.htm) \n\n*   [![Image 9](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2025-12\u002Feebc10f934d74b978fccd8c654f47577.jpg)](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fzyfw\u002Fcqk\u002Findex.htm)\n*   [![Image 10](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002Fcontent\u002F2022-02\u002Faefa6458187845e9a3aeb18472ae9e7f.jpg)](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fzyfw\u002Fcqk\u002Findex.htm)\n*   [![Image 11](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002F5c35b429611f44aa927c9b2698ff91e5.png)](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fzyfw\u002Fcqk\u002Findex.htm)\n*   [![Image 12](https:\u002F\u002Fwww.caas.cn\u002Fimages\u002F2023-01\u002F2ec7c5b598424a0b801fd6fde5956562.png)](https:\u002F\u002Fwww.caas.cn\u002Fxxfw\u002Fzyfw\u002Fcqk\u002Findex.htm)\n\n*   [网站地图](https:\u002F\u002Fwww.caas.cn\u002Fwzdt\u002Findex.htm)\n*   [联系我们](https:\u002F\u002Fwww.caas.cn\u002Flxwm\u002Findex.htm)\n\n[我要捐赠](https:\u002F\u002Fjjh.caas.cn\u002Fjzzn\u002Fjzfs\u002Findex.htm)\n\n主办：中国农业科学院 承办:中国农业科学院农业信息研究所 地址:北京市海淀区中关村南大街12号 邮编:100081\n\nCopyright@中国农业科学院 京ICP备10039560号-5[_![Image 13](https:\u002F\u002Fwww.caas.cn\u002Fstatic2022\u002Fimages\u002Ficon\\_ba.png)_ 京公网安备11010802048565号](http:\u002F\u002Fwww.beian.gov.cn\u002Fportal\u002FregisterSystemInfo?recordcode=11010802048565)","中国农业科学院","2026-09-09T00:00:00Z","报道",10,false,81,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},22,20,17,14,8,1,"国家级科研机构将可解释AI引入农产品产地溯源，方法新颖、结论可靠，对智慧农业质量监管有示范价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","可解释AI","农产品溯源","光谱检测",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":40,"locked":14,"version":41,"method":42,"evidence":43},"classified","2026-10-22.2","rules",{"candidates":44,"agricultural_evidence":46},[45,51,57,64],{"code":33,"score":21,"supported":46,"title":47,"abstract":48,"tags":50},true,[],[28,49],"artificial intelligence",[28,29],{"code":52,"score":53,"supported":46,"title":54,"abstract":55,"tags":56},"I1",4,[],[27],[27],{"code":58,"score":59,"supported":46,"title":60,"abstract":61,"tags":63},"B2",3,[],[62],"猕猴桃",[],{"code":65,"score":66,"supported":14,"title":67,"abstract":68,"tags":69},"E4",2,[],[],[31,30],[71,72],"农业人工智能 农产品溯源 光谱检测 智慧农业","农业人工智能 农产品溯源","农业人工智能农产品溯源光谱检测智慧农业-2084","2026-09-11T00:04:19.946470Z",{"total":76,"page":22,"page_size":76,"items":77},6,[78,165,238,316,366,424],{"id":79,"title":80,"url":81,"summary":82,"summary_zh":83,"content":8,"source_name":84,"source_url":81,"published_at":85,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":87,"score_detail":88,"sources":94,"tags":96,"primary_topic":99,"secondary_topics":100,"classification":101,"search_phrases":127,"slug":130,"view_count":131,"doi":132,"paper":133,"created_at":164},5080,"Explainable deep learning for Mango leaf disease recognition using AlexNet and grad-CAM","https:\u002F\u002Fdoi.org\u002F10.1177\u002F1088467x261495708","Early and accurate detection of plant diseases is essential for improving crop productivity and supporting sustainable agricultural practices. Mango (Mangifera indica), one of the most economically important tropical fruit crops worldwide, is highly susceptible to several leaf diseases that can significantly reduce yield and fruit quality. This study presents an explainable deep learning framework for automated ten-class mango leaf disease classification using a publicly available dataset containing approximately 5000 images distributed across Anthracnose, Bacterial Canker, Cutting Weevil, Dag Disease, Die Back, Galls, Healthy, Powdery Mildew, Red Rust, and Sooty Mould classes. Unlike conventional AlexNet-based approaches, the proposed framework integrates Bayesian hyperparameter optimization, two-stage fine-tuning, test-time augmentation (TTA), and Grad-CAM-based explainability to improve both predictive performance and interpretability. Experimental results demonstrate that the proposed framework achieves a final test accuracy of 99.2% using eight-view test-time augmentation (TTA), with corresponding precision, recall, and F1-score of approximately 99.1%. The model also achieves micro-average and macro-average ROC–AUC scores of 1.00 on the curated test set, indicating strong discriminative capability across the considered disease categories. Comparative evaluations against MobileNetV2, DenseNet201, ConvNeXt, and LeafNet demonstrate competitive or superior classification performance while maintaining moderate computational complexity and architectural simplicity. Furthermore, Grad-CAM visualizations qualitatively highlight biologically relevant disease symptoms, including lesions, necrotic regions, rust spots, fungal growth, and tissue discoloration, thereby enhancing model interpretability. Although field-scale validation and quantitative explainability assessment remain subjects for future investigation, the proposed framework provides an accurate, computationally efficient, and interpretable approach for automated mango leaf disease diagnosis with strong potential for smart agriculture and precision farming applications.","植物病害的早期准确检测对于提高作物生产力和支持可持续农业实践至关重要。芒果（Mangifera indica）是全球最具经济重要性的热带水果作物之一，极易感染多种叶部病害，这些病害可显著降低产量和果实品质。本研究提出了一种可解释的深度学习框架，用于自动化十类芒果叶部病害分类，所使用的公开数据集包含约5000张图像，分布于炭疽病（Anthracnose）、细菌性溃疡病（Bacterial Canker）、切叶象甲（Cutting Weevil）、枯枝病（Dag Disease）、回枯病（Die Back）、虫瘿（Galls）、健康（Healthy）、白粉病（Powdery Mildew）、红锈病（Red Rust）和煤烟病（Sooty Mould）十个类别。与传统的基于AlexNet的方法不同，所提出的框架集成了贝叶斯超参数优化、两阶段微调、测试时增强（TTA）以及基于Grad-CAM的可解释性，以同时提升预测性能和可解释性。实验结果表明，所提出的框架在使用八视图测试时增强（TTA）时达到了99.2%的最终测试准确率，相应的精确率、召回率和F1分数约为99.1%。该模型在精选测试集上还取得了1.00的微平均和宏平均ROC–AUC分数，表明在所考虑的病害类别间具有较强的判别能力。与MobileNetV2、DenseNet201、ConvNeXt和LeafNet的对比评估表明，该框架在保持适中计算复杂度和架构简洁性的同时，具有竞争性或更优的分类性能。此外，Grad-CAM可视化定性地突出了具有生物学相关性的病害症状，包括病斑、坏死区域、锈斑、真菌生长和组织变色，从而增强了模型的可解释性。尽管田间规模验证和定量可解释性评估仍有待未来研究，但所提出的框架为自动化芒果叶部病害诊断提供了一种准确、计算高效且可解释的方法，在智慧农业和精准农业应用中具有强大潜力。","Intelligent Data Analysis","2026-10-15T00:00:00Z","论文",76,{"impact":89,"substance":90,"depth":19,"authority":91,"freshness":92,"relevant":22,"comment":93},16,21,13,9,"将贝叶斯调参、两阶段微调与Grad-CAM结合用于十类芒果叶病识别，准确率99.2%，方法新颖且可解释性强，对智慧农业植保应用有参考价值。",[95],{"name":84,"url":81},[27,28,29,97,98],"精准农业","芒果病害",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":102},{"candidates":103,"agricultural_evidence":46},[104,110,116,121],{"code":33,"score":89,"supported":46,"title":105,"abstract":107,"tags":109},[106],"deep learning",[108,106],"深度学习",[28,29],{"code":52,"score":111,"supported":46,"title":112,"abstract":113,"tags":115},11,[],[27,97,114],"smart agriculture",[27,97],{"code":58,"score":59,"supported":46,"title":117,"abstract":118,"tags":120},[],[119],"芒果",[],{"code":122,"score":59,"supported":46,"title":123,"abstract":124,"tags":126},"H8",[],[125],"可持续农业",[],[128,129],"芒果叶片病害 AlexNet","Grad-CAM 芒果病害识别","芒果叶片病害AlexNet-5080",0,"10.1177\u002F1088467x261495708",{"doi":132,"openalex_id":134,"authors":135,"venue":84,"cited_by_count":131,"oa_url":8,"card":156,"direction":162,"ingested_from":163},"W7222328160",[136,139,142,144,147,150,153],{"name":137,"orcid":138},"Anay Ghosh","https:\u002F\u002Forcid.org\u002F0009-0006-6237-7905",{"name":140,"orcid":141},"Tawfikur Rahman","https:\u002F\u002Forcid.org\u002F0000-0002-0456-5143",{"name":143,"orcid":8},"Md Moniruzzaman",{"name":145,"orcid":146},"Nibedita Deb","https:\u002F\u002Forcid.org\u002F0000-0002-1579-1138",{"name":148,"orcid":149},"Hanan Abdullah Mengash","https:\u002F\u002Forcid.org\u002F0000-0002-4103-2434",{"name":151,"orcid":152},"Noorlindawaty Md Jizat","https:\u002F\u002Forcid.org\u002F0000-0001-7618-2283",{"name":154,"orcid":155},"Samir Salem Al‐Bawri","https:\u002F\u002Forcid.org\u002F0000-0003-2852-575X",{"tldr":157,"method":158,"finding":159,"direction":160,"opportunity":161},"提出可解释深度学习框架，用AlexNet结合贝叶斯调参、微调和Grad-CAM实现芒果叶病十分类。","AlexNet+贝叶斯超参优化、两阶段微调、八视图TTA、Grad-CAM，约5","测试准确率达99.2%，ROC-AUC为1.00，Grad-CAM能定位病斑等生物相关症状。","农业人工智能与决策模型","可延伸至田间复杂背景验证、可解释性定量评估及轻量化边缘部署研究。","智慧农业 \u002F 农业物联网","openalex","2026-10-16T23:30:08.303796Z",{"id":166,"title":167,"url":168,"summary":169,"summary_zh":170,"content":8,"source_name":171,"source_url":168,"published_at":85,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":172,"score_detail":173,"sources":176,"tags":178,"primary_topic":181,"secondary_topics":182,"classification":183,"search_phrases":217,"slug":220,"view_count":22,"doi":221,"paper":222,"created_at":237},5076,"Trustworthy Explainable AI for Lettuce Growth Stage Prediction Using Sensor Data in Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.20944\u002Fpreprints202610.1452.v1","Background In precision agriculture, especially in hydroponic systems where environmental management affects both production and resource efficiency, interpretable modeling of crop phenological stages is crucial. Typical visualization-based approaches are generally expensive and ecologically dependent. To classify and visualize lettuce growth stages (Seedling, Vegetative Growth, Heading, Bolting and Senescence) using temperature, humidity, total dissolved solids (TDS) and pH, this study proposed XGB-StageVision, a sensor-powered machine learning setup combining supervised and unsupervised learning techniques. Using XG Boost, LightGBM, CatBoost, ExtraTrees, Random Forest and LSTM under an 80\u002F20 train-test split, a dataset of 3,169 unique observations from 70 hydroponically grown lettuce plants was evaluated. Methods and Results XGB-StageVision achieved 99.86% test accuracy, a ROC AUC of 1.00, an MCC of 0.9982 and perfect precision, recall, F1 scores and specificity across all classes, performing competitively with the best baseline models, including LightGBM, which achieved statistically comparable cross validation performance. The unsupervised t-SNE visualization showed clearly distinct structural clusters across growth stages, while SHAP revealed temperature as the strongest predictor (≈75% contribution), followed by TDS. Novel Contributions SHAP, LIME and t-SNE provided global, local and structural interpretability and transparent perspectives on environmental influences regarding plant development. Statistical analysis and 10-fold cross-validation were used to further test the model’s robustness, reliability and generalizability. Conclusion XGB-StageVision provides an IoT-ready, interpretable and imaging-free system for monitoring lettuce growth stages in real time. The framework combines supervised predictive modeling with unsupervised structural visualization to ensure transparency, sustainability and high performance, supporting resource-efficient, data-driven strategic planning in contemporary hydroponic agriculture to lower monitoring costs, optimize inputs and improve yield consistency.","背景 在精准农业中，尤其是在环境管理同时影响产量和资源效率的水培系统中，作物物候阶段的可解释建模至关重要。典型的基于可视化的方法通常成本高昂且依赖生态条件。为了利用温度、湿度、总溶解固体（TDS）和pH对生菜生长阶段（幼苗期、营养生长期、结球期、抽薹期和衰老期）进行分类和可视化，本研究提出了XGB-StageVision，一种由传感器驱动的机器学习方案，结合了监督学习和无监督学习技术。在80\u002F20的训练-测试划分下，使用XGBoost、LightGBM、CatBoost、ExtraTrees、随机森林和LSTM，对来自70株水培生菜的3，169个独特观测值数据集进行了评估。方法与结果 XGB-StageVision达到了99.86%的测试准确率、1.00的ROC AUC、0.9982的MCC，并在所有类别上实现了完美的精确率、召回率、F1分数和特异性，与最佳基线模型表现相当，包括LightGBM，其交叉验证性能在统计上具有可比性。无监督t-SNE可视化显示了各生长阶段之间明显不同的结构聚类，而SHAP揭示温度是最强的预测因子（约75%贡献），其次是TDS。新颖贡献 SHAP、LIME和t-SNE提供了全局、局部和结构可解释性，以及关于环境对植物发育影响的可透明视角。使用统计分析和10折交叉验证进一步检验了模型的稳健性、可靠性和泛化性。结论 XGB-StageVision提供了一种物联网就绪、可解释且无需成像的系统，用于实时监测生菜生长阶段。该框架将监督预测建模与无监督结构可视化相结合，以确保透明度、可持续性和高性能，支持当代水培农业中资源高效、数据驱动的战略规划，以降低监测成本、优化投入并提高产量一致性。","Preprints.org",64,{"impact":174,"substance":18,"depth":19,"authority":76,"freshness":92,"relevant":22,"comment":175},12,"预印本论文，方法新颖、数据规模可观且强调可解释性，但尚未经同行评审，属细分领域技术进展。",[177],{"name":171,"url":168},[27,28,29,179,180],"水培生菜","传感器监测",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":184},{"candidates":185,"agricultural_evidence":46},[186,192,198,205,211],{"code":52,"score":187,"supported":46,"title":188,"abstract":189,"tags":191},15,[114],[97,190],"precision agriculture",[27],{"code":33,"score":21,"supported":46,"title":193,"abstract":194,"tags":197},[],[195,196],"机器学习","machine learning",[28,29],{"code":199,"score":76,"supported":46,"title":200,"abstract":201,"tags":204},"I5",[],[202,203],"物联网","IoT",[],{"code":206,"score":59,"supported":46,"title":207,"abstract":208,"tags":210},"B1",[],[209],"生菜",[],{"code":212,"score":59,"supported":46,"title":213,"abstract":214,"tags":216},"B7",[],[215],"hydroponic",[],[218,219],"XGB-StageVision 生菜 生长阶段","水培生菜 传感器 生长预测","XGB-StageVision生菜生长阶段-5076","10.20944\u002Fpreprints202610.1452.v1",{"doi":221,"openalex_id":223,"authors":224,"venue":171,"cited_by_count":131,"oa_url":168,"card":232,"direction":162,"ingested_from":163},"W7222080602",[225,227,230],{"name":226,"orcid":8},"Muhammad Majid",{"name":228,"orcid":229},"Attia Ghafoor","https:\u002F\u002Forcid.org\u002F0009-0000-3828-6479",{"name":231,"orcid":8},"Amna Munir",{"tldr":233,"method":234,"finding":235,"direction":162,"opportunity":236},"提出XGB-StageVision，用传感器数据与可解释AI预测水培生菜生长阶段。","XGBoost等模型结合SHAP、LIME、t-SNE，基于3169条温湿度TD","测试准确率99.86%，温度贡献约75%，t-SNE显示各生长阶段聚类清晰。","可探索多作物、多环境迁移与低成本传感器部署下的可解释模型泛化能力。","2026-10-16T23:30:08.020788Z",{"id":239,"title":240,"url":241,"summary":242,"summary_zh":243,"content":8,"source_name":244,"source_url":241,"published_at":245,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":246,"score_detail":247,"sources":250,"tags":252,"primary_topic":255,"secondary_topics":261,"classification":263,"search_phrases":285,"slug":288,"view_count":22,"doi":289,"paper":290,"created_at":315},4993,"Effects of within-season temporal aggregation of weather variables on performance and interpretability in rice yield machine learning models","https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41598-026-75646-2","While crop responses are inherently stage-dependent, machine learning yield models often rely on weather variables aggregated over the entire growing season. This study evaluates how within-season temporal aggregation of weather variables affects both predictive performance and interpretability in machine learning-based rice yield modeling. Using CYGMA (Crop Yield Growth Model with Assumptions on Climate and Socioeconomics)-simulated potential yield as a synthetic response variable, we minimized confounding effects from observational noise and management variability, enabling a clearer evaluation of the role of within-season weather aggregation. Daily weather variables were averaged over the rice growing period after dividing each season into multiple equal-length segments, and several machine learning algorithms were trained and evaluated using temporally independent validation. Increasing within-season temporal resolution improved model performance up to a moderate level, after which gains diminished. Interpretation using SHAP (SHapley Additive exPlanations) and response diagnostics consistently indicated that yield sensitivity concentrates in specific windows of the season, with temperature-related effects emerging most strongly early in the season and humidity-related effects later. Finer temporal aggregation sharpened the identification of window-specific nonlinear and threshold-like responses that were obscured under coarse seasonal averages. Overall, moderate within-season aggregation provides a practical balance between predictive skill and mechanistic interpretability, supporting more reliable identification of phenology-relevant meteorological risks for climate-impact assessment and decision support.","尽管作物响应本质上具有阶段依赖性，但机器学习产量模型往往依赖于整个生长季聚合的天气变量。本研究评估了天气变量的季节内时间聚合如何影响基于机器学习的水稻产量建模的预测性能与可解释性。以CYGMA（Crop Yield Growth Model with Assumptions on Climate and Socioeconomics，气候与社会经济假设下的作物产量生长模型）模拟的潜在产量作为合成响应变量，我们最大限度地减少了观测噪声和管理变异性带来的混杂效应，从而能够更清晰地评估季节内天气聚合的作用。将每个生长季划分为多个等长时段后，对水稻生长期内的逐日天气变量进行平均，并采用时间独立验证对多种机器学习算法进行训练和评估。提高季节内时间分辨率可提升模型性能，直至达到中等水平，此后增益递减。利用SHAP（SHapley Additive exPlanations，沙普利加性解释）和响应诊断进行的解释一致表明，产量敏感性集中在季节的特定窗口，温度相关效应在季节早期最为显著，湿度相关效应则在后期出现。更精细的时间聚合增强了对窗口特异性非线性和阈值型响应的识别，而这些响应在粗粒度季节平均下被掩盖。总体而言，中等程度的季节内聚合在预测能力与机制可解释性之间提供了实用平衡，有助于更可靠地识别与物候相关的气象风险，以支持气候影响评估和决策支持。","Scientific Reports","2026-10-13T00:00:00Z",72,{"impact":174,"substance":18,"depth":248,"authority":20,"freshness":21,"relevant":22,"comment":249},18,"该研究通过CYGMA模拟数据系统评估了生长季内气象变量时间聚合对水稻产量机器学习模型性能与可解释性的影响，方法新颖、结论可靠，对农业气象风险识别与决策支持有参考价值，但属学术论文，公共影响范围有限。",[251],{"name":244,"url":241},[27,28,29,253,254],"农业气象","水稻产量预测",{"code":256,"slug":257,"name":258,"path":259,"parent_code":260},"A1","rice","水稻","\u002Ftopic\u002Ffield-crops\u002Frice","A",[262],{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":264},{"candidates":265,"agricultural_evidence":46},[266,272,276,281],{"code":33,"score":267,"supported":46,"title":268,"abstract":269,"tags":271},19,[196],[195,196,270],"decision support",[28,29],{"code":256,"score":20,"supported":46,"title":273,"abstract":274,"tags":275},[257],[258,257],[],{"code":277,"score":22,"supported":14,"title":278,"abstract":279,"tags":280},"H7",[],[],[253],{"code":52,"score":22,"supported":14,"title":282,"abstract":283,"tags":284},[],[],[27],[286,287],"水稻产量 机器学习 气象","SHAP 水稻 产量预测","水稻产量机器学习气象-4993","10.1038\u002Fs41598-026-75646-2",{"doi":289,"openalex_id":291,"authors":292,"venue":244,"cited_by_count":131,"oa_url":309,"card":310,"direction":162,"ingested_from":163},"W7221657053",[293,295,297,299,301,303,306],{"name":294,"orcid":8},"Noh-Hyun Lee",{"name":296,"orcid":8},"Wonjae Jeong",{"name":298,"orcid":8},"Hyeon-Ji Yang",{"name":300,"orcid":8},"Hana Cho",{"name":302,"orcid":8},"Hyeon-Yeong Ju",{"name":304,"orcid":305},"Toshichika Iizumi","https:\u002F\u002Forcid.org\u002F0000-0002-0611-4637",{"name":307,"orcid":308},"Ki‐Tae Kim","https:\u002F\u002Forcid.org\u002F0000-0002-8249-5283","https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41598-026-75646-2_reference.pdf",{"tldr":311,"method":312,"finding":313,"direction":160,"opportunity":314},"评估生长季内天气变量时间聚合粒度对水稻产量机器学习模型预测与可解释性的影响。","用CYGMA模拟潜在产量作响应变量，按等长时段聚合逐日天气，训练多种机器学习模型","适度提高季内时间分辨率可提升性能，产量敏感性集中在特定窗口，温度效应偏早、湿度效应偏晚。","可探索不同作物与气候区下最优聚合粒度，并将窗口特异性气象风险嵌入气候影响评估与决策支持系统。","2026-10-15T23:30:10.483881Z",{"id":317,"title":318,"url":319,"summary":320,"summary_zh":321,"content":8,"source_name":322,"source_url":319,"published_at":245,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":172,"score_detail":323,"sources":325,"tags":327,"primary_topic":330,"secondary_topics":331,"classification":332,"search_phrases":349,"slug":352,"view_count":59,"doi":353,"paper":354,"created_at":365},4907,"PEAS-DeiT: Multi-Scale Spatial Attention Enhanced DeiT for Pea Leaf Disease Classification","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fapp162010113","Early and accurate detection of pea diseases is essential for reducing crop losses and supporting precision agriculture. Although deep learning methods have shown good performance in plant disease classification, CNN-based models may have limitations in learning long-range relationships and detailed disease features. In this study, an explainable transformer-based framework, named PEAS-DeiT, is proposed for automated pea leaf disease classification. The proposed approach integrates a multi-scale spatial attention mechanism into the Data-efficient Image Transformer (DeiT) architecture to enhance the representation of disease-related visual features. A unified pea leaf disease dataset (UPLDD) was constructed by combining multiple publicly available datasets, covering six classes, including five disease categories and healthy samples. The proposed model was evaluated against CNN-based and transformer-based architectures using five-fold cross-validation. Furthermore, explainable artificial intelligence (XAI) analysis was performed to investigate the regions contributing to model predictions. Experimental results demonstrate that PEAS-DeiT achieved an accuracy of 93.11%, showing competitive performance compared with the evaluated CNN and transformer-based models. The gradient-based saliency analysis also revealed localized response patterns in several evaluated samples, providing qualitative insight into the image regions contributing to model predictions. Statistical analyses were conducted to evaluate the performance differences among the compared models. These findings indicate that PEAS-DeiT provides a competitive approach for pea leaf disease classification within the evaluated experimental setting.","豌豆病害的早期准确检测对于减少作物损失和支持精准农业至关重要。尽管深度学习方法在植物病害分类中表现出良好性能，但基于卷积神经网络（CNN）的模型在学习长程关系和病害细节特征方面可能存在局限。本研究提出了一种基于Transformer的可解释框架PEAS-DeiT，用于豌豆叶片病害的自动化分类。该方法将多尺度空间注意力机制集成到数据高效图像Transformer（DeiT）架构中，以增强病害相关视觉特征的表征能力。通过整合多个公开数据集，构建了统一的豌豆叶片病害数据集（UPLDD），涵盖六个类别，包括五类病害和健康样本。采用五折交叉验证，将所提模型与基于CNN和基于Transformer的架构进行了评估比较。此外，进行了可解释人工智能（XAI）分析，以探究对模型预测有贡献的区域。实验结果表明，PEAS-DeiT达到了93.11%的准确率，与所评估的基于CNN和基于Transformer的模型相比展现出有竞争力的性能。基于梯度的显著性分析还揭示了若干评估样本中的局部响应模式，为理解对模型预测有贡献的图像区域提供了定性见解。通过统计分析评估了所比较模型之间的性能差异。这些发现表明，在所评估的实验设置下，PEAS-DeiT为豌豆叶片病害分类提供了一种有竞争力的方法。","Applied Sciences",{"impact":21,"substance":248,"depth":89,"authority":91,"freshness":92,"relevant":22,"comment":324},"提出多尺度空间注意力增强的DeiT模型用于豌豆叶病害分类，构建六类统一数据集并达93.11%准确率，方法有创新但属实验室验证阶段，产业影响有限。",[326],{"name":322,"url":319},[27,28,29,328,329],"植物病害识别","豌豆",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":333},{"candidates":334,"agricultural_evidence":46},[335,339,344],{"code":33,"score":174,"supported":46,"title":336,"abstract":337,"tags":338},[],[108,106,49],[28,29,328],{"code":52,"score":340,"supported":46,"title":341,"abstract":342,"tags":343},7,[],[97,190],[27],{"code":345,"score":22,"supported":14,"title":346,"abstract":347,"tags":348},"A5",[],[],[329],[350,351],"PEAS-DeiT 豌豆叶病害","DeiT 植物病害分类","PEAS-DeiT豌豆叶病害-4907","10.3390\u002Fapp162010113",{"doi":353,"openalex_id":355,"authors":356,"venue":322,"cited_by_count":131,"oa_url":319,"card":360,"direction":160,"ingested_from":163},"W7221612639",[357],{"name":358,"orcid":359},"Fatma Ozge Ozkok","https:\u002F\u002Forcid.org\u002F0000-0002-1421-4670",{"tldr":361,"method":362,"finding":363,"direction":160,"opportunity":364},"提出PEAS-DeiT模型，将多尺度空间注意力融入DeiT，实现豌豆叶片病害分类。","多尺度空间注意力增强DeiT，构建UPLDD数据集，五折交叉验证与XAI分析。","PEAS-DeiT达到93.11%准确率，优于对比的CNN和Transformer模型。","可探索轻量化Transformer在移动端病害识别中的部署，并扩展多作物跨域泛化能力。","2026-10-14T23:30:46.091813Z",{"id":367,"title":368,"url":369,"summary":370,"summary_zh":371,"content":8,"source_name":372,"source_url":369,"published_at":373,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":374,"score_detail":375,"sources":377,"tags":379,"primary_topic":382,"secondary_topics":383,"classification":384,"search_phrases":406,"slug":409,"view_count":22,"doi":410,"paper":411,"created_at":423},4833,"From Handcrafted Descriptors to Deep Embeddings: A Comparative Study for Regional Hazelnut Image Classification","https:\u002F\u002Fdoi.org\u002F10.15832\u002Fankutbd.1946462","Automated regional hazelnut image classification is relevant for agricultural traceability, origin aware sorting, and decision support in food processing systems. This study presents a unified comparative benchmark for three class hazelnut image classification using Van, Ordu, and Giresun categories. The experimental framework integrates classical handcrafted descriptors, deep transfer learning models, and reduced space hybrid classical and quantum compatible classifiers. In the classical branch, HOG and RGB color histogram features were combined and evaluated using KNN and SVM based classifiers. In the deep learning branch, ResNet50 and EfficientNetB0 were trained through a two stage transfer learning strategy, and Grad-CAM was employed to provide qualitative interpretability. In the hybrid branch, 1280 dimensional embeddings extracted from the GAP layer of the best EfficientNetB0 model were compressed into low dimensional spaces for reduced space SVC and QSVC experiments. Under the present image level split protocol, the strongest handcrafted baseline, PCA + SVM, achieved 98.90% accuracy and 0.9888 macro-F1, while ResNet50 reached 99.45% accuracy and 0.9945 macro-F1. EfficientNetB0 achieved 100% test accuracy, while the reduced space classical SVC trained on compressed EfficientNetB0 embeddings attained 99.63% accuracy and 0.9961 macro F1 on the full test set. Under the fair subset protocol, classical SVC achieved 98.33% accuracy, whereas the best QSVC configuration reached 90.83% accuracy. This result should be interpreted as a strong within dataset benchmark rather than definitive evidence of object level or deployment level generalization. Grad-CAM visualizations further suggested that the deep models focused primarily on semantically meaningful hazelnut regions rather than on background patterns. The findings indicate that representation quality is a central factor in this task, while the reported near ceiling performance should be interpreted in light of the image level partitioning strategy","区域榛子图像的自动化分类对于农业可追溯性、产地感知分选以及食品加工系统中的决策支持具有重要意义。本研究提出了一个统一的比较基准，用于对Van、Ordu和Giresun三个类别的榛子图像进行分类。实验框架整合了经典手工设计描述子、深度迁移学习模型以及降维空间下的经典与量子兼容混合分类器。在经典分支中，将HOG与RGB颜色直方图特征进行组合，并使用基于KNN和SVM的分类器进行评估。在深度学习分支中，通过两阶段迁移学习策略训练ResNet50和EfficientNetB0，并采用Grad-CAM提供定性可解释性。在混合分支中，将从最佳EfficientNetB0模型GAP层提取的1280维嵌入压缩至低维空间，用于降维空间SVC和QSVC实验。在当前图像级划分协议下，最强的手工特征基线PCA + SVM达到了98.90%的准确率和0.9888的宏F1值，而ResNet50达到了99.45%的准确率和0.9945的宏F1值。EfficientNetB0达到了100%的测试准确率，而在压缩后的EfficientNetB0嵌入上训练的降维空间经典SVC在完整测试集上达到了99.63%的准确率和0.9961的宏F1值。在公平子集协议下，经典SVC达到了98.33%的准确率，而最佳QSVC配置达到了90.83%的准确率。该结果应被解释为数据集内的强基准，而非对象级或部署级泛化能力的确定性证据。Grad-CAM可视化进一步表明，深度模型主要关注具有语义意义的榛子区域，而非背景模式。研究结果表明，表征质量是该任务的核心因素，而所报告的接近上限的性能应结合图像级划分策略来理解。","Tarım Bilimleri Dergisi","2026-10-12T00:00:00Z",63,{"impact":21,"substance":248,"depth":89,"authority":174,"freshness":92,"relevant":22,"comment":376},"方法对比扎实、结论可靠，但属数据集内基准研究，产业落地价值有限，可作技术参考而非每日精选头条。",[378],{"name":372,"url":369},[27,28,30,380,381],"图像识别","榛子",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":385},{"candidates":386,"agricultural_evidence":46},[387,391,398,402],{"code":33,"score":111,"supported":46,"title":388,"abstract":389,"tags":390},[],[108,106,270],[28,380],{"code":392,"score":76,"supported":46,"title":393,"abstract":394,"tags":397},"E1",[],[395,396],"食品加工","food processing",[],{"code":65,"score":22,"supported":14,"title":399,"abstract":400,"tags":401},[],[],[30],{"code":52,"score":22,"supported":14,"title":403,"abstract":404,"tags":405},[],[],[27],[407,408],"榛子 产地 图像分类","EfficientNetB0 榛子 识别","榛子产地图像分类-4833","10.15832\u002Fankutbd.1946462",{"doi":410,"openalex_id":412,"authors":413,"venue":372,"cited_by_count":131,"oa_url":417,"card":418,"direction":160,"ingested_from":163},"W7221535543",[414],{"name":415,"orcid":416},"Özgür Tomak","https:\u002F\u002Forcid.org\u002F0000-0003-2993-6913","https:\u002F\u002Fdergipark.org.tr\u002Fen\u002Fdownload\u002Farticle-file\u002F5990175",{"tldr":419,"method":420,"finding":421,"direction":160,"opportunity":422},"比较手工特征、深度迁移学习和量子兼容分类器对三个地区榛子图像的分类性能。","HOG+颜色直方图、ResNet50\u002FEfficientNetB0迁移学习、Gr","EfficientNetB0达100%准确率，深度嵌入质量是关键，但结果受图像级划分限制。","可探索对象级划分与跨域泛化，并验证量子分类器在真实农业溯源场景的实用性。","2026-10-13T23:30:38.487402Z",{"id":425,"title":426,"url":427,"summary":428,"summary_zh":429,"content":8,"source_name":430,"source_url":427,"published_at":373,"category":86,"cover_url":8,"hotness":13,"is_selected":14,"score":431,"score_detail":432,"sources":434,"tags":436,"primary_topic":438,"secondary_topics":439,"classification":440,"search_phrases":471,"slug":474,"view_count":131,"doi":475,"paper":476,"created_at":500},4811,"Explainable deep learning framework for real-time insect pest detection in smart agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44163-026-02109-y","Abstract Global losses of crop production due to insect pest infestations are estimated to be around 20–40% annually, which represents one of the highest threats to agricultural productivity, farmer livelihoods and food security. Current manual scouting is labor intensive, subjective and non-scalable, and many automated scouting methods suffer when they come to deal with different field conditions and offer little insight into decision making. To tackle these challenges, this study proposes Crop Protection in Smart Agriculture—Real-time Insect Pest Detection using Continual Spatio-Temporal Graph Convolutional Network (CPSA-RIPD-CSTGCN), a unified and interpretable deep learning framework that combines the four synergistic components: (i) the Ensemble Transform Schmidt–Kalman Filter (ETSKF) for image preprocessing, including resizing, normalization, and noise suppression; (ii) data augmentation using rotation, flipping, scaling, shearing, and contrast adjustment, which enhances generalization and reduces overfitting; (iii) the Continual Spatio-Temporal Graph Convolutional Network (CSTGCN) that jointly models the spatial dependency across crop regions and temporal evolution of pest dynamics to achieve robust feature extraction and classification; and (iv) the Revolution Optimization Algorithm (ROA) for hyperparameter optimization of CSTGCN weights to adaptively adjust the network. Further, Gradient-weighted Class Activation Mapping (Grad-CAM) visualization is introduced to give post-hoc explainability as it shows regions that are relevant to the biological process and areas that are affected by pests. This framework was tested on IP102 benchmark dataset with 60%-40%-split train-test with 10 economically important classes of pests with 97% accuracy, 98% precision, 98% recall and 98% F1 score, mean ± 0.32% accuracy, across 5-fold cross validation. Among other baseline models, these results consistently outperform VGG16 (82%), InceptionV3 (85%), MobileNetV2 (87%), DenseNet121 (88%), EfficientNetB0 (89%) and task-specific models like EDCP-DIA-RNN(Early Disease and Crop Pest—Diagnosis\u002FDetection using Image Analysis RNN) (94%), EPCY-TPA-ANN (Efficient Prediction of Crop Yield—Through Parameter Analysis ANN) (91%) and ICPM-ESA-CNN (Interactive Crop and Pest Management Enhanced Soil Analysis CNN) (90%). The results of ablation studies showed the solitary effects of each module of the Ensemble Transform Schmidt–Kalman Filter pre-processing and Revolution Optimization Algorithm optimization, and the confusion matrix analysis demonstrated high per-class reliability and only a small amount of misclassification, implying that the proposed method could be applied to real-time, large-scale smart agriculture deployments.","摘要 由于害虫侵扰导致的全球作物产量损失估计每年约为20–40%，这是对农业生产力、农民生计和粮食安全的最大威胁之一。当前的人工田间巡查劳动强度大、主观性强且难以规模化，而许多自动化巡查方法在面对不同田间条件时表现不佳，且对决策制定提供的洞见有限。为应对这些挑战，本研究提出了智慧农业中的作物保护——基于持续时空图卷积网络的实时害虫检测（CPSA-RIPD-CSTGCN），这是一个统一且可解释的深度学习框架，融合了四个协同组件：（i）集成变换施密特-卡尔曼滤波器（ETSKF），用于图像预处理，包括调整尺寸、归一化和噪声抑制；（ii）使用旋转、翻转、缩放、剪切和对比度调整进行数据增强，以增强泛化能力并减少过拟合；（iii）持续时空图卷积网络（CSTGCN），联合建模作物区域间的空间依赖性和害虫动态的时间演化，以实现鲁棒的特征提取和分类；（iv）革命优化算法（ROA），用于CSTGCN权重的超参数优化，以自适应调整网络。此外，引入梯度加权类激活映射（Grad-CAM）可视化以提供事后可解释性，因为它显示了与生物过程相关的区域以及受害虫影响的区域。该框架在IP102基准数据集上进行了测试，采用60%-40%的训练-测试划分，涵盖10个具有经济重要性的害虫类别，在5折交叉验证中达到了97%的准确率、98%的精确率、98%的召回率和98%的F1分数，准确率均值±0.32%。在其他基线模型中，这些结果持续优于VGG16（82%）、InceptionV3（85%）、MobileNetV2（87%）、DenseNet121（88%）、EfficientNetB0（89%）以及特定任务模型，如EDCP-DIA-RNN（基于图像分析的早期病害和作物害虫诊断\u002F检测RNN）（94%）、EPCY-TPA-ANN（通过参数分析人工神经网络高效预测作物产量）（91%）和ICPM-ESA-CNN（交互式作物和害虫管理增强土壤分析CNN）（90%）。消融研究的结果显示了集成变换施密特-卡尔曼滤波器预处理和革命优化算法各模块的单独效果。","Discover Artificial Intelligence",80,{"impact":248,"substance":17,"depth":248,"authority":91,"freshness":92,"relevant":22,"comment":433},"提出可解释时空图卷积虫害实时检测框架，在IP102基准上达97%准确率并优于多种主流模型，方法新颖、数据扎实，对智慧农业植保具有参考价值。",[435],{"name":430,"url":427},[27,28,29,108,437],"病虫害检测",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":441},{"candidates":442,"agricultural_evidence":46},[443,447,451,458,465],{"code":33,"score":19,"supported":46,"title":444,"abstract":445,"tags":446},[106],[108,106],[28,29,108,437],{"code":52,"score":187,"supported":46,"title":448,"abstract":449,"tags":450},[114],[27,114],[27],{"code":452,"score":76,"supported":46,"title":453,"abstract":454,"tags":457},"A8",[],[455,456],"粮食安全","food security",[],{"code":459,"score":76,"supported":46,"title":460,"abstract":461,"tags":464},"H2",[],[462,463],"土壤","soil",[],{"code":466,"score":59,"supported":46,"title":467,"abstract":468,"tags":470},"F4",[],[469],"crop protection",[],[472,473],"IP102 害虫检测","CSTGCN 虫害识别","IP102害虫检测-4811","10.1007\u002Fs44163-026-02109-y",{"doi":475,"openalex_id":477,"authors":478,"venue":430,"cited_by_count":131,"oa_url":427,"card":495,"direction":162,"ingested_from":163},"W7221565168",[479,482,485,488,491,493],{"name":480,"orcid":481},"J. Briskilal","https:\u002F\u002Forcid.org\u002F0000-0003-4849-3249",{"name":483,"orcid":484},"M. K. Vidhyalakshmi","https:\u002F\u002Forcid.org\u002F0000-0002-6879-9897",{"name":486,"orcid":487},"Aswathy K Cherian","https:\u002F\u002Forcid.org\u002F0009-0002-3679-2583",{"name":489,"orcid":490},"S. Neduncheliyan","https:\u002F\u002Forcid.org\u002F0009-0002-0904-8695",{"name":492,"orcid":8},"Ajay Kumar",{"name":494,"orcid":8},"Rakesh Kumar",{"tldr":496,"method":497,"finding":498,"direction":160,"opportunity":499},"提出可解释深度学习框架CPSA-RIPD-CSTGCN，实现智慧农业中害虫实时检测。","用CSTGCN时空图卷积、ROA超参优化、ETSKF预处理与Grad-CAM，在","10类害虫检测准确率97%，优于VGG16、EfficientNetB0及多种任务专用模型。","可探索轻量化模型在边缘设备部署，并融合多模态田间数据提升跨域泛化能力。","2026-10-13T23:30:10.529316Z"]