[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-176":3,"related-176":79},{"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":73,"slug":76,"view_count":77,"doi":8,"paper":8,"created_at":78},176,"Information fusion in smart agriculture: machine learning applications and future research directions","https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002Fs1566253525011029","综述系统分析 ML 在农业中的应用，强调多源数据融合（遥感、IoT、生物数据）对精准农业的关键作用。围绕四个研究目标展开：（1）探索 ML 在收获前、收获中、收获后阶段的整合；（2）展示多源数据融合的益处；（3）研究当前研究趋势；（4）汇编关键公共数据集。重点提及 Transformer、GAN、联邦学习、持续学习和可解释 AI 等新兴方法在应对现代农业挑战中的作用。",null,"[![Image 1: Elsevier logo](blob:http:\u002F\u002Flocalhost\u002F84fae110a9934890163c7653d951a57a)](https:\u002F\u002Fwww.sciencedirect.com\u002F)\n\n*   Help   \n\n# Are you a robot?\n\nPlease confirm you are a human by completing the captcha challenge below.\n\nVerification successful. Waiting for www.sciencedirect.com to respond\n\n*   **Reference number:**a248c563af7b2fe3\n*   **IP Address:**34.96.49.178\n*   **User Agent:**Mozilla\u002F5.0 (X11; Linux x86_64) AppleWebKit\u002F537.36 (KHTML, like Gecko) Chrome\u002F151.0.0.0 Safari\u002F537.36\n*   **Timestamp:**2026-08-01 23:54:59 UTC\n\n[![Image 2: Elsevier logo with wordmark](blob:http:\u002F\u002Flocalhost\u002F59eaca882375907228cfe15f113faae5)](https:\u002F\u002Fwww.elsevier.com\u002F)\n\n*   [About ScienceDirect](https:\u002F\u002Fwww.elsevier.com\u002Fsolutions\u002Fsciencedirect)\n*   [Remote access](https:\u002F\u002Fwww.sciencedirect.com\u002Fuser\u002Finstitution\u002Flogin?targetURL=%2F)\n*   [Contact and support](https:\u002F\u002Fservice.elsevier.com\u002Fapp\u002Fcontact\u002Fsupporthub\u002Fsciencedirect\u002F)\n*   [Terms and conditions](https:\u002F\u002Fwww.elsevier.com\u002Flegal\u002Felsevier-website-terms-and-conditions)\n*   [Privacy policy](https:\u002F\u002Fwww.elsevier.com\u002Flegal\u002Fprivacy-policy)\n*   Cookie settings\n\nAll content on this site: Copyright © 2026 Elsevier B.V., its licensors, and contributors. All rights are reserved, including those for text and data mining, AI training, and similar technologies. For all open access content, the relevant licensing terms apply.\n\n[![Image 3: RELX group home page](blob:http:\u002F\u002Flocalhost\u002F5f7d1b1a12ba08c5b8ef9e17a8f9a957)](https:\u002F\u002Fwww.relx.com\u002F)\n\n\",promise:n.resolve()}}}},\"core\u002Fsrc\u002Flib\u002Factions\u002Fhelpers\u002Fdecorators\u002FdecorateNonGlobalJavaScriptCode.js\":{script:function(e,t,a){\"use strict\";var n=a(\"@adobe\u002Freactor-promise\"),r=0;e.exports=function(e,t){var a=\"_runScript\"+ ++r,i=new n((function(t,r){_satellite[a]=function(i){delete _satellite[a],new n((function(t){t(i.call(e.event.element,e.event,e.event.target,n))})).then(t,r)}}));return{code:'\",promise:i}}}},\"core\u002Fsrc\u002Flib\u002Factions\u002Fhelpers\u002Fdecorators\u002FdecorateHtmlCode.js\":{script:function(e,t,a,n){\"use strict\";var r=a(\"@adobe\u002Freactor-promise\"),i=0,s={};window._satellite=window._satellite||{},window._satellite._onCustomCodeSuccess=function(e){var t=s[e];t&&(delete s[e],t.resolve())},window._satellite._onCustomCodeFailure=function(e){var t=s[e];t&&(delete s[e],t.reject())};var o=function(e){return-1!==e.indexOf(\"${reactorCallbackId}\")},c=function(e,t){return e.replace(\u002F\\${reactorCallbackId}\u002Fg,t)},l=function(e){return e.settings.isExternal};e.exports=function(e,t){var a;return l(e)&&(t=n.replaceTokens(t,e.event)),o(t)?(a=new r((function(e,t){s[String(i)]={resolve:e,reject:t}})),t=c(t,i),i+=1):a=r.resolve(),{code:t,promise:a}}}},\"core\u002Fsrc\u002Flib\u002Factions\u002Fhelpers\u002FgetSourceByUrl.js\":{script:function(e,t,a){\"use strict\";var n=a(\"@adobe\u002Freactor-load-script\"),r=a(\"@adobe\u002Freactor-promise\"),i=a(\"..\u002F..\u002Fhelpers\u002FfindPageScript\").byRegexPattern,s={},o={},c=function(e){return o[e]||(o[e]=n(e)),o[e]};_satellite.__registerScript=function(e,t){var a;if(document.currentScript)a=document.currentScript.getAttribute(\"src\");else{var n=new RegExp(\".*\"+e+\".*\");a=i(n).getAttribute(\"src\")}s[a]=t},e.exports=function(e){return s[e]?r.resolve(s[e]):new r((function(t){c(e).then((function(){t(s[e])}),(function(){t()}))}))}}},\"core\u002Fsrc\u002Flib\u002Fevents\u002Fhelpers\u002FpageLifecycleEvents.js\":{script:function(e,t,a){\"use strict\";var n=a(\"@adobe\u002Freactor-window\"),r=a(\"@adobe\u002Freactor-document\"),i=-1!==n.navigator.appVersion.indexOf(\"MSIE 10\"),s=\"WINDOW_LOADED\",o=\"DOM_READY\",c=\"PAGE_BOTTOM\",l=[c,o,s],u=function(e,t){return{element:e,target:e,nativeEvent:t}},d={};l.forEach((function(e){d[e]=[]}));var p=function(e,t){l.slice(0,v(e)+1).forEach((function(e){g(t,e)}))},m=function(){return\"complete\"===r.readyState?s:\"interactive\"===r.readyState?i?null:o:void 0},v=function(e){return l.indexOf(e)},g=function(e,t){d[t].forEach((function(t){f(e,t)})),d[t]=[]},f=function(e,t){var a=t.trigger,n=t.syntheticEventFn;a(n?n(e):null)};n._satellite=n._satellite||{},n._satellite.pageBottom=p.bind(null,c),r.addEventListener(\"DOMContentLoaded\",p.bind(null,o),!0),n.addEventListener(\"load\",p.bind(null,s),!0),n.setTimeout((function(){var e=m();e&&p(e)}),0),e.exports={registerLibraryLoadedTrigger:function(e){e()},registerPageBottomTrigger:function(e){d[c].push({trigger:e})},registerDomReadyTrigger:function(e){d[o].push({trigger:e,syntheticEventFn:u.bind(null,r)})},registerWindowLoadedTrigger:function(e){d[s].push({trigger:e,syntheticEventFn:u.bind(null,n)})}}}}}},\"adobe-mcid\":{displayName:\"Experience Cloud ID Service\",hostedLibFilesBaseUrl:\"https:\u002F\u002Fassets.adobedtm.com\u002Fextensions\u002FEP31a59fd25d824db7be52972a70e94c1c\u002F\",settings:{orgId:\"4D6368F454EC41940A4C98A6@AdobeOrg\",variables:[{name:\"serverState\",value:\"%serverState%\"},{name:\"trackingServer\",value:\"metrics.elsevier.com\"},{name:\"trackingServerSecure\",value:\"smetrics.elsevier.com\"},{name:\"marketingCloudServer\",value:\"metrics.elsevier.com\"},{name:\"marketingCloudServerSecure\",value:\"smetrics.elsevier.com\"},{name:\"loadSSL\",value:\"true\"},{name:\"overwriteCrossDomainMCIDAndAID\",value:\"%mcmid - pass params%\"}]},modules:{\"adobe-mcid\u002Fsrc\u002Flib\u002FsharedModules\u002FmcidInstance.js\":{script:function(e,t,a,n){\"use strict\";var r=a(\"@adobe\u002Freactor-document\"),i=a(\"..\u002FcodeLibrary\u002FVisitorAPI\"),s=a(\"..\u002F..\u002Fview\u002Futils\u002FtimeUnits\"),o=function(e){return e.reduce((function(e,t){var a=\u002F^(true|false)$\u002Fi.test(t.value)?JSON.parse(t.value):t.value;return e[t.name]=a,e}),{})},c=function(e){var t=n.getExtensionSettings();if(\"string\"!=typeof t.orgId)throw new TypeError(\"Org ID is not a string.\");var a=o(t.variables||[]),r=t.doesOptInApply;r&&(\"boolean\"==typeof r?a.doesOptInApply=r:t.optInCallback&&(a.doesOptInApply=t.optInCallback));var i=t.isOptInStorageEnabled;i&&(a.isOptInStorageEnabled=i);var c=t.optInCookieDomain;c&&(a.optInCookieDomain=c);var l=t.optInStorageExpiry;if(l){var u=t.timeUnit;if(u&&s[u]){var d=l*s[u];a.optInStorageExpiry=d}}else!0===i&&(a.optInStorageExpiry=33696e3);var p=t.previousPermissions;p&&(a.previousPermissions=p);var m=t.preOptInApprovals;if(m)a.preOptInApprovals=m;else{var v=t.preOptInApprovalInput;v&&(a.preOptInApprovals=v)}var g=t.isIabContext;g&&(a.isIabContext=g);var f=e.getInstance(t.orgId,a);return n.logger.info('Created instance using orgId: \"'+t.orgId+'\"'),n.logger.info(\"Set variables: \"+JSON.stringify(a)),f.getMarketingCloudVisitorID((function(e){n.logger.info(\"Obtained Marketing Cloud Visitor Id: \"+e)}),!0),f},l=function(e){return(n.getExtensionSettings().pathExclusions||[]).some((function(t){return t.valueIsRegex?new RegExp(t.value,\"i\").test(e):t.value===e}))},u=null;_satellite.getVisitorId=function(){return u},l(r.location.pathname)?n.logger.warn(\"MCID library not loaded. One of the path exclusions matches the current path.\"):u=c(i),e.exports=u},name:\"mcid-instance\",shared:!0},\"adobe-mcid\u002Fsrc\u002Flib\u002FcodeLibrary\u002FVisitorAPI.js\":{script:function(e){e.exports=(function(){\"use strict\";function e(t){return(e=\"function\"==typeof Symbol&&\"symbol\"==typeof Symbol.iterator?function(e){return typeof e}:function(e){return e&&\"function\"==typeof Symbol&&e.constructor===Symbol&&e!==Symbol.prototype?\"symbol\":typeof e})(t)}function t(e,t,a){return t in e?Object.defineProperty(e,t,{value:a,enumerable:!0,configurable:!0,writable:!0}):e[t]=a,e}function a(){return{callbacks:{},add:function(e,t){this.callbacks[e]=this.callbacks[e]||[];var a=this.callbacks[e].push(t)-1,n=this;return function(){n.callbacks[e].splice(a,1)}},execute:function(e,t){if(this.callbacks[e]){t=(t=void 0===t?[]:t)instanceof Array?t:[t];try{for(;this.callbacks[e].length;){var a=this.callbacks[e].shift();\"function\"==typeof a?a.apply(null,t):a instanceof Array&&a[1].apply(a[0],t)}delete this.callbacks[e]}catch(e){}}},executeAll:function(e,t){(t||e&&!A.isObjectEmpty(e))&&Object.keys(this.callbacks).forEach((function(t){var a=void 0!==e[t]?e[t]:\"\";this.execute(t,a)}),this)},hasCallbacks:function(){return Boolean(Object.keys(this.callbacks).length)}}}function n(e,t,a){var n=null==e?void 0:e[t];return void 0===n?a:n}function r(e){for(var t=\u002F^\\d+$\u002F,a=0,n=e.length;a r)return 1;if(r>n)return-1}return 0}function o(e,t){if(e===t)return 0;var a=e.toString().split(\".\"),n=t.toString().split(\".\");return r(a.concat(n))?(i(a,n),s(a,n)):NaN}function c(e){return e===Object(e)&&0===Object.keys(e).length}function l(e){return\"function\"==typeof e||e instanceof Array&&e.length}function u(){var e=arguments.length>0&&void 0!==arguments[0]?arguments[0]:\"\",t=arguments.length>1&&void 0!==arguments[1]?arguments[1]:function(){return!0};this.log=ue(\"log\",e,t),this.warn=ue(\"warn\",e,t),this.error=ue(\"error\",e,t)}function d(){var e=(arguments.length>0&&void 0!==arguments[0]?arguments[0]:{}).cookieName,t=(arguments.length>1&&void 0!==arguments[1]?arguments[1]:{}).cookies;if(!e||!t)return{get:Ce,set:Ce,remove:Ce};var a={remove:function(){t.remove(e)},get:function(){var a=t.get(e),n={};try{n=JSON.parse(a)}catch(a){n={}}return n},set:function(n,r){r=r||{};var i=a.get(),s=Object.assign(i,n);t.set(e,JSON.stringify(s),{domain:r.optInCookieDomain||\"\",cookieLifetime:r.optInStorageExpiry||3419e4,secure:r.secure,sameSite:r.sameSite,expires:!0})}};return a}function p(e){this.name=this.constructor.name,this.message=e,\"function\"==typeof Error.captureStackTrace?Error.captureStackTrace(this,this.constructor):this.stack=new Error(e).stack}function m(){function e(e,t){var a=ge(e);return a.length?a.every((function(e){return!!t[e]})):fe(t)}function t(){w(T),S(te.COMPLETE),b(h.status,h.permissions),o&&f.set(h.permissions,{optInCookieDomain:c,optInStorageExpiry:l,secure:p,sameSite:m}),y.execute(je)}function a(e){return function(a,n){if(!he(a))throw new Error(\"[OptIn] Invalid category(-ies). Please use the `OptIn.Categories` enum.\");return S(te.CHANGED),Object.assign(T,be(ge(a),e)),n||t(),h}}var n=arguments.length>0&&void 0!==arguments[0]?arguments[0]:{},r=n.doesOptInApply,i=n.previousPermissions,s=n.preOptInApprovals,o=n.isOptInStorageEnabled,c=n.optInCookieDomain,l=n.optInStorageExpiry,u=n.isIabContext,p=n.secureCookie,m=n.sameSiteCookie,v=(arguments.length>1&&void 0!==arguments[1]?arguments[1]:{}).cookies,g=Te(i);Se(g,\"Invalid `previousPermissions`!\"),Se(s,\"Invalid `preOptInApprovals`!\");var f=d({cookieName:\"adobeujs-optin\"},{cookies:v}),h=this,b=ee(h),y=oe(),k=De(g),D=De(s),V=o?f.get():{},P={},E=function(e,t){return Ve(e)||t&&Ve(t)?te.COMPLETE:te.PENDING}(k,V),C=function(e,t,a){var n=be(se,!r);return r?Object.assign({},n,e,t,a):n}(D,k,V),T=ye(C),S=function(e){return E=e},w=function(e){return C=e};h.deny=a(!1),h.approve=a(!0),h.denyAll=h.deny.bind(h,se),h.approveAll=h.approve.bind(h,se),h.isApproved=function(t){return e(t,h.permissions)},h.isPreApproved=function(t){return e(t,D)},h.fetchPermissions=function(e){var t=arguments.length>1&&void 0!==arguments[1]&&arguments[1],a=t?h.on(te.COMPLETE,e):Ce;return!r||r&&h.isComplete||s?e(h.permissions):t||y.add(je,(function(){return e(h.permissions)})),a},h.complete=function(){h.status===te.CHANGED&&t()},h.registerPlugin=function(e){if(!e||!e.name||\"function\"!=typeof e.onRegister)throw new Error(_e);P[e.name]||(P[e.name]=e,e.onRegister.call(e,h))},h.execute=Ae(P),h.memoizeContent=function(e){Ee(e)&&f.set(e,{optInCookieDomain:c,optInStorageExpiry:l,secure:p,sameSite:m})},h.getMemoizedContent=function(e){var t=f.get();if(t)return t[e]},Object.defineProperties(h,{permissions:{get:function(){return C}},status:{get:function(){return E}},Categories:{get:function(){return ae}},doesOptInApply:{get:function(){return!!r}},isPending:{get:function(){return h.status===te.PENDING}},isComplete:{get:function(){return h.status===te.COMPLETE}},__plugins:{get:function(){return Object.keys(P)}},isIabContext:{get:function(){return u}}})}function v(e,t){function a(){r=null,e.call(e,new p(\"The call took longer than you wanted!\"))}function n(){r&&(clearTimeout(r),e.apply(e,arguments))}if(void 0===t)return e;var r=setTimeout(a,t);return n}function g(){if(window.__tcfapi)return window.__tcfapi;var e=window;if(e!==window.top){for(var t;!t;){e=e.parent;try{e.frames.__tcfapiLocator&&(t=e)}catch(e){}if(e===window.top)break}if(t){var a={};return window.__tcfapi=function(e,n,r,i){var s=Math.random()+\"\",o={__tcfapiCall:{command:e,parameter:i,version:n,callId:s}};a[s]=r,t.postMessage(o,\"*\")},window.addEventListener(\"message\",(function(e){var t=e.data;if(\"string\"==typeof t)try{t=JSON.parse(e.data)}catch(e){}if(t.__tcfapiReturn){var n=t.__tcfapiReturn;\"function\"==typeof a[n.callId]&&(a[n.callId](n.returnValue,n.success),delete a[n.callId])}}),!1),window.__tcfapi}me.error(\"__tcfapi not found\")}else me.error(\"__tcfapi not found\")}function f(e,t){var a=arguments.length>2&&void 0!==arguments[2]?arguments[2]:[],n=!0===e.vendor.consents[t],r=a.every((function(t){return!0===e.purpose.consents[t]}));return n&&r}function h(){var e=this;e.name=\"iabPlugin\",e.version=\"0.0.2\";var t,a=oe(),n={transparencyAndConsentData:null},r=function(e){var t=arguments.length>1&&void 0!==arguments[1]?arguments[1]:{};return n[e]=t};e.fetchConsentData=function(e){var t=v(e.callback,e.timeout);i({callback:t})},e.isApproved=function(e){var t=e.callback,a=e.category,r=e.timeout;if(n.transparencyAndConsentData)return t(null,f(n.transparencyAndConsentData,ne[a],re[a]));var s=v((function(e,n){t(e,f(n,ne[a],re[a]))}),r);i({category:a,callback:s})},e.onRegister=function(a){t=a;var n=Object.keys(ne),r=function(e,t){!e&&t&&(n.forEach((function(e){var n=f(t,ne[e],re[e]);a[n?\"approve\":\"deny\"](e,!0)})),a.complete())};e.fetchConsentData({callback:r})};var i=function(e){var i=e.callback;if(n.transparencyAndConsentData)return i(null,n.transparencyAndConsentData);a.add(\"FETCH_CONSENT_DATA\",i),s((function(e,i){if(i){var s=ye(e),o=t.getMemoizedContent(\"iabConsentHash\"),c=pe(s.tcString).toString(32);s.consentString=e.tcString,s.hasConsentChangedSinceLastCmpPull=o!==c,r(\"transparencyAndConsentData\",s),t.memoizeContent({iabConsentHash:c})}a.execute(\"FETCH_CONSENT_DATA\",[null,n.transparencyAndConsentData])}))},s=function(e){var t=Ie(ne),a=g();\"function\"==typeof a&&a(\"getTCData\",2,e,t)}}var b=\"undefined\"!=typeof globalThis?globalThis:\"undefined\"!=typeof window?window:\"undefined\"!=typeof global?global:\"undefined\"!=typeof self?self:{};Object.assign=Object.assign||function(e){for(var t,a,n=1;n=0||t.indexOf(\"Trident\u002F\")>=0&&t.indexOf(\"Windows NT 6\")>=0};a.getIeVersion=function(){return document.documentMode?document.documentMode:n()?7:null},a.isFirefox=function(e){return!!\u002FFirefox\\\u002F([0-9\\.]+)(?:\\s|$)\u002F.test(e||window.navigator.userAgent)},a.encodeAndBuildRequest=function(e,t){return e.map(encodeURIComponent).join(t)},a.isObject=function(t){return null!==t&&\"object\"===e(t)&&!1===Array.isArray(t)},a.defineGlobalNamespace=function(){return window.adobe=a.isObject(window.adobe)?window.adobe:{},window.adobe},a.pluck=function(e,t){return t.reduce((function(t,a){return e[a]&&(t[a]=e[a]),t}),Object.create(null))},a.parseOptOut=function(e,t,a){t||(t=a,e.d_optout&&e.d_optout instanceof Array&&(t=e.d_optout.join(\",\")));var n=parseInt(e.d_ottl,10);return isNaN(n)&&(n=7200),{optOut:t,d_ottl:n}},a.normalizeBoolean=function(e){var t=e;return\"true\"===e?t=!0:\"false\"===e&&(t=!1),t}})),j=(A.isObjectEmpty,A.isValueEmpty,A.getIeVersion,A.isFirefox,A.encodeAndBuildRequest,A.isObject,A.defineGlobalNamespace,A.pluck,A.parseOptOut,A.normalizeBoolean,a),_=D.MESSAGES,N={0:\"prefix\",1:\"orgID\",2:\"state\"},O=function(e,t){this.parse=function(e){try{var t={};return e.data.split(\"|\").forEach((function(e,a){void 0!==e&&(t[N[a]]=2!==a?e:JSON.parse(e))})),t}catch(e){}},this.isInvalid=function(a){var n=this.parse(a);if(!n||Object.keys(n).length\u003C2)return!0;var r=e!==n.orgID,i=!t||a.origin!==t,s=-1===Object.keys(_).indexOf(n.prefix);return r||i||s},this.send=function(a,n,r){var i=n+\"|\"+e;r&&r===Object(r)&&(i+=\"|\"+JSON.stringify(r));try{a.postMessage(i,t)}catch(e){}}},L=D.MESSAGES,x=function(e,t,a,n){function r(e){Object.assign(m,e)}function i(e){Object.assign(m.state,e),Object.assign(m.state.ALLFIELDS,e),m.callbackRegistry.executeAll(m.state)}function s(e){if(!f.isInvalid(e)){g=!1;var t=f.parse(e);m.setStateAndPublish(t.state)}}function o(e){!g&&v&&(g=!0,f.send(n,e))}function c(){r(new P(a._generateID)),m.getMarketingCloudVisitorID(),m.callbackRegistry.executeAll(m.state,!0),b.removeEventListener(\"message\",l)}function l(e){if(!f.isInvalid(e)){var t=f.parse(e);g=!1,b.clearTimeout(m._handshakeTimeout),b.removeEventListener(\"message\",l),r(new S(m)),b.addEventListener(\"message\",s),m.setStateAndPublish(t.state),m.callbackRegistry.hasCallbacks()&&o(L.GETSTATE)}}function u(){v&&postMessage?(b.addEventListener(\"message\",l),o(L.HANDSHAKE),m._handshakeTimeout=setTimeout(c,250)):c()}function d(){b.s_c_in||(b.s_c_il=[],b.s_c_in=0),m._c=\"Visitor\",m._il=b.s_c_il,m._in=b.s_c_in,m._il[m._in]=m,b.s_c_in++}function p(){function e(e){0!==e.indexOf(\"_\")&&\"function\"==typeof a[e]&&(m[e]=function(){})}Object.keys(a).forEach(e),m.getSupplementalDataID=a.getSupplementalDataID,m.isAllowed=function(){return!0}}var m=this,v=t.whitelistParentDomain;m.state={ALLFIELDS:{}},m.version=a.version,m.marketingCloudOrgID=e,m.cookieDomain=a.cookieDomain||\"\",m._instanceType=\"child\";var g=!1,f=new O(e,v);m.callbackRegistry=j(),m.init=function(){d(),p(),r(new I(m)),u()},m.findField=function(e,t){if(void 0!==m.state[e])return t(m.state[e]),m.state[e]},m.messageParent=o,m.setStateAndPublish=i},M=D.MESSAGES,R=D.ALL_APIS,F=D.ASYNC_API_MAP,B=D.FIELDGROUP_TO_FIELD,U=function(e,t){function a(){var t={};return Object.keys(R).forEach((function(a){var n=R[a],r=e[n]();A.isValueEmpty(r)||(t[a]=r)})),t}function n(){var t=[];return e._loading&&Object.keys(e._loading).forEach((function(a){if(e._loading[a]){var n=B[a];t.push(n)}})),t.length?t:null}function r(t){return function a(){var r=n();if(r){var i=F[r[0]];e[i](a,!0)}else t()}}function i(e,n){var r=a();t.send(e,n,r)}function s(e){c(e),i(e,M.HANDSHAKE)}function o(e){r((function(){i(e,M.PARENTSTATE)}))()}function c(a){function n(n){r.call(e,n),t.send(a,M.PARENTSTATE,{CUSTOMERIDS:e.getCustomerIDs()})}var r=e.setCustomerIDs;e.setCustomerIDs=n}return function(e){t.isInvalid(e)||(t.parse(e).prefix===M.HANDSHAKE?s:o)(e.source)}},H=function(e,t){function a(e){return function(a){n[e]=a,++r===i&&t(n)}}var n={},r=0,i=Object.keys(e).length;Object.keys(e).forEach((function(t){var n=e[t];if(n.fn){var r=n.args||[];r.unshift(a(t)),n.fn.apply(n.context||null,r)}}))},W={get:function(e){e=encodeURIComponent(e);var t=(\";\"+document.cookie).split(\" \").join(\";\"),a=t.indexOf(\";\"+e+\"=\"),n=a\u003C0?a:t.indexOf(\";\",a+1);return a\u003C0?\"\":decodeURIComponent(t.substring(a+2+e.length,n\u003C0?t.length:n))},set:function(e,t,a){var r=n(a,\"cookieLifetime\"),i=n(a,\"expires\"),s=n(a,\"domain\"),o=n(a,\"secure\"),c=n(a,\"sameSite\"),l=o?\"Secure\":\"\",u=c?\"SameSite=\"+c+\";\":\"\";if(i&&\"SESSION\"!==r&&\"NONE\"!==r){var d=\"\"!==t?parseInt(r||0,10):-60;if(d)(i=new Date).setTime(i.getTime()+1e3*d);else if(1===i){var p=(i=new Date).getYear();i.setYear(p+2+(p\u003C1900?1900:0))}}else i=0;return e&&\"NONE\"!==r?(document.cookie=encodeURIComponent(e)+\"=\"+encodeURIComponent(t)+\"; path=\u002F;\"+(i?\" expires=\"+i.toGMTString()+\";\":\"\")+(s?\" domain=\"+s+\";\":\"\")+u+l,this.get(e)===t):0},remove:function(e,t){var a=n(t,\"domain\");a=a?\" domain=\"+a+\";\":\"\";var r=n(t,\"secure\"),i=n(t,\"sameSite\"),s=r?\"Secure\":\"\",o=i?\"SameSite=\"+i+\";\":\"\";document.cookie=encodeURIComponent(e)+\"=; Path=\u002F; Expires=Thu, 01 Jan 1970 00:00:01 GMT;\"+a+o+s}},G=function(e,t){!e&&b.location&&(e=b.location.hostname);var a,n=e.split(\".\"),r=t||{};for(a=n.length-2;a>=0;a--)if(r.domain=n.slice(a).join(\".\"),W.set(\"TEST_AMCV_COOKIE_WRITE\",\"cookie\",r))return W.remove(\"TEST_AMCV_COOKIE_WRITE\",r),r.domain;return\"\"},Y={compare:o,isLessThan:function(e,t){return o(e,t)\u003C0},areVersionsDifferent:function(e,t){return 0!==o(e,t)},isGreaterThan:function(e,t){return o(e,t)>0},isEqual:function(e,t){return 0===o(e,t)}},q=!!b.postMessage,X={postMessage:function(e,t,a){var n=1;t&&(q?a.postMessage(e,t.replace(\u002F([^:]+:\\\u002F\\\u002F[^\\\u002F]+).*\u002F,\"$1\")):t&&(a.location=t.replace(\u002F#.*$\u002F,\"\")+\"#\"+ +new Date+n+++\"&\"+e))},receiveMessage:function(e,t){var a;try{q&&(e&&(a=function(a){if(\"string\"==typeof t&&a.origin!==t||\"[object Function]\"===Object.prototype.toString.call(t)&&!1===","Information Fusion Vol. 129, May 2026, 104040","2026-05-15T00:00:00Z","论文",10,false,78,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},25,20,18,15,0,1,"核心期刊综述，主题高度相关，但正文被反爬拦截，无法获取实质内容，时效性差。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","机器学习","可解释AI","信息融合","联邦学习",{"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":48},[45,54,60,67],{"code":46,"score":47,"supported":48,"title":49,"abstract":51,"tags":53},"I1",12,true,[50],"smart agriculture",[52],"精准农业",[27],{"code":33,"score":55,"supported":48,"title":56,"abstract":58,"tags":59},11,[57],"machine learning",[],[29,28,31],{"code":61,"score":62,"supported":48,"title":63,"abstract":64,"tags":66},"I3",3,[],[65],"遥感",[],{"code":68,"score":62,"supported":48,"title":69,"abstract":70,"tags":72},"I5",[],[71],"IoT",[],[74,75],"信息融合 智慧农业 机器学习 联邦学习","信息融合 智慧农业","信息融合智慧农业机器学习联邦学习-176",5,"2026-08-01T23:53:44.566287Z",{"total":80,"page":22,"page_size":80,"items":81},6,[82,171,264,353,428,482],{"id":83,"title":84,"url":85,"summary":86,"summary_zh":87,"content":8,"source_name":88,"source_url":85,"published_at":89,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":90,"score_detail":91,"sources":97,"tags":99,"primary_topic":102,"secondary_topics":103,"classification":104,"search_phrases":150,"slug":153,"view_count":77,"doi":154,"paper":155,"created_at":170},3626,"A Hybrid Model for Crop Yield Prediction Using Recurrent Neural Networks and Explainable Artificial Intelligence","https:\u002F\u002Fdoi.org\u002F10.22194\u002Fjgias\u002F27.2061","Accurate maize yield prediction is essential for ensuring food security and supporting agricultural planning in Kenya. However, the changes in climate and severe weather are posing more challenges to the stability of yield and food security. While advanced machine learning models, such as Long Short–Term Memory (LSTM) networks, offer strong predictive performance, their inherent \"black–box\" nature limits their practical adoption by farmers and policymakers who require interpretable and trustworthy insights. This study developed a transparent predictive model by integrating LSTM with Explainable Artificial Intelligence (XAI). Using county–level maize yield data (2012–2023), the hybrid model was compared against Random Forest and Gradient Boosting baselines. The LSTM model achieved superior performance (R² = 0.8551, RMSE = 0.3715, MAE = 0.2705), compared to Random Forest (R2 = 0.8441, RMSE= 0.3851 and MAE= 0.2856) and Gradient Boosting (R2 = 0.7925, RMSE = 0.4443 and MAE = 0.3234) baselines. XAI analysis, using SHapley Additive explanations (SHAP) and Local Interpretable Model–agnostic Explanations (LIME), identified longitude, latitude, and annual rainfall as key predictors. The model maintained high accuracy while improving interpretability, increasing transparency and trust. This provides actionable insights for farmers and policymakers, supporting evidence–based planning and enhancing resilience in smallholder systems. Keywords: Crop yield prediction, long short–term memory (LSTM), explainable artificial intelligence, precision agriculture, machine learning.","准确的玉米产量预测对于保障肯尼亚粮食安全和支持农业规划至关重要。然而，气候变化和极端天气对产量稳定性和粮食安全构成了更多挑战。尽管长短期记忆（LSTM）网络等先进机器学习模型具有强大的预测性能，但其固有的“黑箱”特性限制了农民和政策制定者的实际采用，因为他们需要可解释且可信的洞见。本研究通过将LSTM与可解释人工智能（XAI）相结合，开发了一个透明的预测模型。利用县级玉米产量数据（2012—2023年），将该混合模型与随机森林和梯度提升基线模型进行了比较。LSTM模型取得了更优的性能（R² = 0.8551，RMSE = 0.3715，MAE = 0.2705），优于随机森林（R2 = 0.8441，RMSE= 0.3851，MAE= 0.2856）和梯度提升（R2 = 0.7925，RMSE = 0.4443，MAE = 0.3234）基线模型。使用SHapley加法解释（SHAP）和局部可解释模型无关解释（LIME）进行的XAI分析，确定了经度、纬度和年降雨量是关键预测因子。该模型在保持高精度的同时提高了可解释性，增强了透明度和信任。这为农民和政策制定者提供了可操作的洞见，支持基于证据的规划，并增强小农系统的韧性。关键词：作物产量预测，长短期记忆（LSTM），可解释人工智能，精准农业，机器学习。","Journal of Global Innovations in Agricultural Sciences","2026-09-25T00:00:00Z",71,{"impact":47,"substance":92,"depth":93,"authority":94,"freshness":95,"relevant":22,"comment":96},21,17,13,8,"将LSTM与可解释AI结合用于肯尼亚县级玉米产量预测，方法新颖、指标详实，对智慧农业有参考价值，但属境外区域研究，公共影响有限。",[98],{"name":88,"url":85},[27,100,28,29,101],"农业人工智能","玉米产量预测",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":105},{"candidates":106,"agricultural_evidence":48},[107,112,118,125,132,138,144],{"code":33,"score":18,"supported":48,"title":108,"abstract":110,"tags":111},[109],"artificial intelligence",[28,57,109],[100,29,28],{"code":46,"score":113,"supported":48,"title":114,"abstract":115,"tags":117},7,[],[52,116],"precision agriculture",[27],{"code":119,"score":80,"supported":48,"title":120,"abstract":121,"tags":124},"A3",[],[122,123],"玉米","maize",[],{"code":126,"score":80,"supported":48,"title":127,"abstract":128,"tags":131},"A8",[],[129,130],"粮食安全","food security",[],{"code":133,"score":62,"supported":48,"title":134,"abstract":135,"tags":137},"G2",[],[136],"smallholder",[],{"code":139,"score":62,"supported":48,"title":140,"abstract":141,"tags":143},"H7",[],[142],"气候变化",[],{"code":145,"score":62,"supported":48,"title":146,"abstract":147,"tags":149},"J1",[],[148],"农业规划",[],[151,152],"LSTM 玉米产量预测","SHAP LIME 农业模型","LSTM玉米产量预测-3626","10.22194\u002Fjgias\u002F27.2061",{"doi":154,"openalex_id":156,"authors":157,"venue":88,"cited_by_count":21,"oa_url":85,"card":163,"direction":167,"ingested_from":169},"W7214446524",[158,160],{"name":159,"orcid":8},"Stephen Gitau Ndung’u",{"name":161,"orcid":162},"Consolata Gakii","https:\u002F\u002Forcid.org\u002F0000-0003-2783-9992",{"tldr":164,"method":165,"finding":166,"direction":167,"opportunity":168},"用LSTM结合可解释AI预测肯尼亚玉米产量，兼顾精度与透明度。","基于2012-2023县级玉米产量数据，LSTM融合SHAP与LIME，对比随机","LSTM精度最高（R²=0.8551），经度、纬度和年降雨量是主要预测因子。","农业人工智能与决策模型","可探索将可解释AI与遥感、气象多源数据融合，提升小农户区域产量预测的可信度与推广性。","openalex","2026-09-27T23:31:22.771605Z",{"id":172,"title":173,"url":174,"summary":175,"summary_zh":176,"content":8,"source_name":177,"source_url":174,"published_at":178,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":179,"score_detail":180,"sources":183,"tags":185,"primary_topic":189,"secondary_topics":195,"classification":203,"search_phrases":239,"slug":242,"view_count":21,"doi":243,"paper":244,"created_at":263},5558,"Estimation of Soil Water Content in Winter Wheat Field at Jointing Stage Using Multi-Source Features and Machine Learning","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagronomy16202071","Accurate and timely monitoring of soil water content (SWC) plays an essential supporting role in crop irrigation management and smart agricultural practice. Remote sensing technology provides rapid and convenient approaches to large-scale farmland SWC estimation or inversion. However, existing mechanistic modeling methods are limited by parameter localization and land surface heterogeneity, while machine learning data-driven methods lack quite good physical interpretability. To address deficiencies in single data-driven methods and single vegetation microwave scattering models, and explore the estimation effect of different feature variables in machine learning. This study proposes an innovative estimation model that integrates machine learning strategies and multi-source features, including spectral indices and features derived from the classic water cloud model (WCM). The novelty of the proposed model lies in balancing physical interpretability and data-driven accuracy, namely extracting relevant soil backscattering parameters through WCM, which are then used as input features in the machine learning. A combination of field-measured data and unmanned aerial vehicle (UAV) multi-source remote sensing data, including multispectral and synthetic aperture radar (SAR) data, was acquired synchronously at the jointing stage of winter wheat. Through extracting variables such as the SAR microwave backscattering coefficient and the optical vegetation index, and conducting feature screening and parameter sensitivity analysis, we optimized the input variable sets of models based on back-propagation (BP) neural network, support vector regression (SVR) and random forest (RF) algorithms. The study results showed that the established estimation model had excellent estimation performance, especially for the RF algorithm, with the highest R2 of 0.855, lowest MAE of 0.0212 cm3\u002Fcm3, and lowest RMSE of 0.0253 cm3\u002Fcm3. It was also shown that different feature input schemes have obvious effects on SWC estimation performance. On the whole, RF had the best estimation accuracy, followed by SVR and then BPNN. This study demonstrated that models integrating multi-source remote sensing data and machine learning could enhance SWC estimation ability for winter wheat fields, and could also serve as a promising approach for SWC monitoring in winter wheat fields in the North China Plain.","准确、及时地监测土壤含水量（SWC）对作物灌溉管理和智慧农业实践具有重要的支撑作用。遥感技术为大规模农田土壤含水量估算或反演提供了快速便捷的手段。然而，现有机理建模方法受参数本地化和地表异质性限制，机器学习数据驱动方法则缺乏良好的物理解释性。针对单一数据驱动方法和单一植被微波散射模型的不足，并探究不同特征变量在机器学习中的估算效果，本研究提出了一种融合机器学习策略与多源特征的创新估算模型，所采用特征包括光谱指数和经典水云模型（WCM）衍生的特征。该模型的新颖性在于平衡物理解释性与数据驱动精度，即通过WCM提取相关土壤后向散射参数，再将其作为机器学习的输入特征。研究在冬小麦拔节期同步获取了田间实测数据与无人机（UAV）多源遥感数据，包括多光谱和合成孔径雷达（SAR）数据。通过提取SAR微波后向散射系数和光学植被指数等变量，并进行特征筛选和参数敏感性分析，优化了基于反向传播（BP）神经网络、支持向量回归（SVR）和随机森林（RF）算法的模型输入变量集。研究结果表明，所建立的估算模型具有优良的估算性能，尤其是RF算法，最高R²为0.855，最低MAE为0.0212 cm³\u002Fcm³，最低RMSE为0.0253 cm³\u002Fcm³。研究还表明，不同特征输入方案对土壤含水量估算性能有明显影响。总体而言，RF估算精度最佳，SVR次之，BPNN再次之。本研究证明，融合多源遥感数据与机器学习的模型能够提升冬小麦田土壤含水量估算能力，也可为华北平原冬小麦田土壤含水量监测提供一种有前景的方法。","Agronomy","2026-10-21T00:00:00Z",77,{"impact":181,"substance":92,"depth":93,"authority":94,"freshness":13,"relevant":22,"comment":182},16,"该研究融合多源遥感与机器学习估算冬小麦拔节期土壤含水量，方法有创新、结论可靠，对智慧灌溉有参考价值，但属细分领域论文，影响范围有限。",[184],{"name":177,"url":174},[27,28,186,187,188],"遥感监测","土壤墒情","冬小麦",{"code":190,"slug":191,"name":192,"path":193,"parent_code":194},"A2","wheat","小麦","\u002Ftopic\u002Ffield-crops\u002Fwheat","A",[196,202],{"code":197,"slug":198,"name":199,"path":200,"parent_code":201},"H2","soil","土壤健康与改良","\u002Ftopic\u002Fenvironment\u002Fsoil","H",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},{"status":40,"locked":14,"version":41,"method":42,"evidence":204},{"candidates":205,"agricultural_evidence":48},[206,212,217,221,227,234],{"code":190,"score":207,"supported":48,"title":208,"abstract":210,"tags":211},26,[209,191],"winter wheat",[192,209,191],[188],{"code":197,"score":20,"supported":48,"title":213,"abstract":214,"tags":216},[198],[215,198],"土壤",[187],{"code":33,"score":20,"supported":48,"title":218,"abstract":219,"tags":220},[57],[28,57],[28],{"code":61,"score":13,"supported":48,"title":222,"abstract":223,"tags":226},[],[65,224,225],"remote sensing","multispectral",[186],{"code":228,"score":80,"supported":48,"title":229,"abstract":230,"tags":233},"H3",[],[231,232],"灌溉","irrigation",[],{"code":46,"score":235,"supported":48,"title":236,"abstract":237,"tags":238},4,[],[27],[27],[240,241],"冬小麦 土壤含水量 遥感","多源遥感 机器学习 墒情","冬小麦土壤含水量遥感-5558","10.3390\u002Fagronomy16202071",{"doi":243,"openalex_id":245,"authors":246,"venue":177,"cited_by_count":21,"oa_url":174,"card":256,"direction":262,"ingested_from":169},"W7223156204",[247,250,252,254],{"name":248,"orcid":249},"Mingliang Jiang","https:\u002F\u002Forcid.org\u002F0000-0002-6486-6087",{"name":251,"orcid":8},"Yi Zhang",{"name":253,"orcid":8},"WEI Lei",{"name":255,"orcid":8},"Chengcai Zhang",{"tldr":257,"method":258,"finding":259,"direction":260,"opportunity":261},"融合WCM物理特征与多源遥感，用机器学习估算冬小麦拔节期土壤含水量。","无人机多光谱与SAR数据，WCM参数+光谱指数，BP\u002FSVR\u002FRF建模。","RF最优，R²达0.855，RMSE为0.0253 cm³\u002Fcm³，特征方案影响显著。","农业遥感与作物表型","可探索物理约束与深度学习耦合，并迁移至多生育期、多作物及不同土壤质地验证。","智慧农业 \u002F 农业物联网","2026-10-22T11:20:18.887301Z",{"id":265,"title":266,"url":267,"summary":268,"summary_zh":269,"content":8,"source_name":270,"source_url":267,"published_at":271,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":272,"score_detail":273,"sources":277,"tags":279,"primary_topic":282,"secondary_topics":283,"classification":290,"search_phrases":326,"slug":329,"view_count":21,"doi":330,"paper":331,"created_at":352},5400,"Artificial intelligence in food safety: a bibliometric and empirical synthesis of emerging technologies and research frontiers","https:\u002F\u002Fdoi.org\u002F10.1108\u002Fbfj-06-2025-0819","Purpose This study aimed to identify recent research trends in the integration of artificial intelligence (AI) with food safety and security through a comprehensive bibliometric analysis. Design\u002Fmethodology\u002Fapproach A total of 2,782 documents published between 2021 and 2025 were retrieved from the Web of Science Core Collection. The dataset was refined by removing duplicates, filtering document types and excluding irrelevant records. VOSviewer was then used to conduct co-authorship, keyword co-occurrence, citation, co-citation and bibliographic coupling analyses. Findings Machine learning and deep learning emerged as core technologies associated with food security, contamination detection and quality prediction. Keyword clustering demonstrated interdisciplinary connections with agriculture, climate change and public health. Citation and co-citation analyses identified three major intellectual foundations: traditional machine learning, computer vision-based deep learning and time-series modeling. China, India and the United States were the most influential contributors in terms of research output and technological development. Bibliographic coupling and source co-citation analyses further indicated increasing convergence between food science and computational science. Originality\u002Fvalue This study provides a comprehensive overview of the intellectual structure, collaborative networks and thematic evolution of AI-driven food safety research. Unlike previous studies focusing on specific technologies or applications, it maps global research patterns and interdisciplinary linkages. The findings provide strategic insights for researchers, policymakers and industry stakeholders. Future studies should prioritize explainable AI integration into regulatory frameworks and the development of larger, more diverse datasets to improve model generalizability across food systems.","目的 本研究旨在通过全面的文献计量分析，识别人工智能（AI）与食品安全及保障融合领域的最新研究趋势。设计与方法 从Web of Science核心合集中检索了2021年至2025年间发表的2，782篇文献。通过去除重复记录、筛选文献类型并排除不相关记录对数据集进行精炼。随后使用VOSviewer进行合著、关键词共现、引文、共被引及文献耦合分析。研究发现 机器学习和深度学习成为与食品安全、污染检测和质量预测相关的核心技术。关键词聚类显示了与农业、气候变化和公共卫生的跨学科联系。引文和共被引分析识别出三大主要知识基础：传统机器学习、基于计算机视觉的深度学习和时间序列建模。中国、印度和美国在研究产出和技术发展方面是最具影响力的贡献者。文献耦合和来源共被引分析进一步表明食品科学与计算科学之间的融合日益加深。原创性\u002F价值 本研究全面概述了AI驱动的食品安全研究的知识结构、合作网络和主题演变。与以往聚焦特定技术或应用的研究不同，本研究绘制了全球研究模式和跨学科联系。研究结果为研究人员、政策制定者和行业利益相关者提供了战略性见解。未来研究应优先关注可解释AI融入监管框架以及开发更大、更多样化的数据集，以提高模型在食品系统中的泛化能力。","British Food Journal","2026-10-19T00:00:00Z",80,{"impact":19,"substance":274,"depth":19,"authority":275,"freshness":95,"relevant":22,"comment":276},22,14,"基于2782篇文献的AI食品安全计量分析，方法规范、结论有参考价值，但属综述性研究，产业落地影响有限。",[278],{"name":270,"url":267},[27,100,28,280,281],"食品安全","文献计量",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[284],{"code":285,"slug":286,"name":287,"path":288,"parent_code":289},"E4","food-safety","质量安全与溯源","\u002Ftopic\u002Ffood-supply\u002Ffood-safety","E",{"status":40,"locked":14,"version":41,"method":42,"evidence":291},{"candidates":292,"agricultural_evidence":48},[293,302,307,313,318,322],{"code":33,"score":294,"supported":48,"title":295,"abstract":296,"tags":301},19,[109],[297,28,298,57,299,109,300],"深度学习","计算机视觉","deep learning","computer vision",[100,28],{"code":285,"score":20,"supported":48,"title":303,"abstract":305,"tags":306},[304],"food safety",[280,304],[280],{"code":308,"score":20,"supported":48,"title":309,"abstract":311,"tags":312},"I7",[310],"bibliometric",[281,310],[281],{"code":139,"score":80,"supported":48,"title":314,"abstract":315,"tags":317},[],[142,316],"climate change",[],{"code":126,"score":62,"supported":48,"title":319,"abstract":320,"tags":321},[],[130],[],{"code":46,"score":22,"supported":14,"title":323,"abstract":324,"tags":325},[],[],[27],[327,328],"AI 食品安全 文献计量","机器学习 食品质量预测","AI食品安全文献计量-5400","10.1108\u002Fbfj-06-2025-0819",{"doi":330,"openalex_id":332,"authors":333,"venue":270,"cited_by_count":21,"oa_url":8,"card":347,"direction":167,"ingested_from":169},"W7222920742",[334,336,338,341,344],{"name":335,"orcid":8},"Harang Lee",{"name":337,"orcid":8},"지은서",{"name":339,"orcid":340},"Yoonbin Kim","https:\u002F\u002Forcid.org\u002F0000-0002-5625-2560",{"name":342,"orcid":343},"Joon-Goo Lee","https:\u002F\u002Forcid.org\u002F0000-0003-3617-5518",{"name":345,"orcid":346},"Woo‐Ju Kim","https:\u002F\u002Forcid.org\u002F0000-0002-6954-1580",{"tldr":348,"method":349,"finding":350,"direction":167,"opportunity":351},"对2021-2025年2782篇文献做文献计量，梳理AI与食品安全交叉研究趋势。","Web of Science文献计量，VOSviewer共被引、关键词共现与耦合","机器学习与深度学习是核心，形成传统ML、计算机视觉、时序建模三大知识基础。","可解释AI嵌入监管框架、构建更大更多样数据集以提升跨食物系统泛化能力。","2026-10-20T23:30:31.567802Z",{"id":354,"title":355,"url":356,"summary":357,"summary_zh":358,"content":8,"source_name":359,"source_url":356,"published_at":360,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":179,"score_detail":361,"sources":363,"tags":365,"primary_topic":367,"secondary_topics":368,"classification":369,"search_phrases":385,"slug":388,"view_count":21,"doi":389,"paper":390,"created_at":427},5306,"A growing challenge: stakeholder perceptions of AI and ML applications in U.S. agriculture","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11119-026-10459-x","Abstract Context The global population is expected to exceed ten billion by 2050 and food demand is projected to increase by up to 100% in the same timeframe. Agricultural management must adapt to meet the increasing global demands. Artificial intelligence (AI) and machine learning (ML) technologies can help the agricultural community respond to these needs. Aims An online survey and follow-up interviews were conducted to examine patterns of use, perceptions, and barriers among early adopting stakeholders regarding AI and ML. Methods The online survey with agricultural stakeholders (n=129) included questions regarding the participants’ views on incorporating AI and ML techniques into farming practices and individual follow-up interviews with agricultural stakeholders (n=10) were conducted to better understand their concerns toward technology use and adoption of AI-enabled tools and technologies. Key results Using a mixed-methods approach, four challenges to establishing AI-\u002FML-enabled technology in agriculture were identified. Namely, challenges related to (1) systems of systems (SoS), (2) sub-seasonal to seasonal weather (S2S) forecasts, (3) precision soil management, and (4) policy and economics. Each challenge was discussed in the context of recent literature. Conclusion Concerns were voiced clustered around four broad research themes. This brief communication brings the perspective of agricultural stakeholders to the discussion of the potential of AI\u002FML-enabled technology and the direction of their development in the agricultural sector. Implications Data collected in 2021, establish an important baseline for the early phase of AI\u002FML use in U.S. agriculture, enabling comparison of future perceptions and developments in the sector. Impact This brief communication describes a survey and follow-up interviews conducted with agricultural stakeholders on factors influencing their willingness and reluctance to incorporate AI and ML techniques into farming practices. The findings, summarized and discussed as four key challenges for AI and ML in the agricultural sector, have the potential to streamline research activities.","摘要\n\n背景 预计到2050年全球人口将超过100亿，同期粮食需求预计将增加多达100%。农业管理必须作出调整以满足不断增长的全球需求。人工智能（AI）和机器学习（ML）技术可帮助农业界应对这些需求。\n\n目的 通过开展在线调查和后续访谈，考察早期采用相关技术的利益相关者在AI和ML方面的使用模式、认知及障碍。\n\n方法 针对农业利益相关者（n=129）的在线调查包含有关参与者对将AI和ML技术纳入农业实践看法的问题，并对农业利益相关者（n=10）进行个体后续访谈，以更好地了解他们对技术使用及采用AI赋能工具和技术的顾虑。\n\n主要结果 采用混合方法，识别出在农业中建立AI\u002FML赋能技术面临的四项挑战，即与以下方面相关的挑战：（1）系统之系统（SoS），（2）次季节至季节天气（S2S）预报，（3）精准土壤管理，以及（4）政策与经济。每项挑战均结合近期文献进行了讨论。\n\n结论 所表达的顾虑集中于四个广泛的研究主题。本简报将农业利益相关者的视角引入关于AI\u002FML赋能技术潜力及其在农业部门发展方向讨论中。\n\n意义 2021年收集的数据为美国农业中AI\u002FML应用的早期阶段建立了重要基线，有助于比较该部门未来的认知和发展。\n\n影响 本简报描述了针对农业利益相关者开展的一项调查和后续访谈，内容涉及影响其愿意和不愿将AI和ML技术纳入农业实践的因素。研究结果总结并讨论为农业部门AI和ML的四项关键挑战，有望优化研究活动。","Precision Agriculture","2026-10-18T00:00:00Z",{"impact":19,"substance":18,"depth":93,"authority":275,"freshness":95,"relevant":22,"comment":362},"基于129份问卷与10次访谈的混合方法研究，系统梳理美国农业AI\u002FML应用的四大挑战，为早期采用者认知提供基线数据，对智慧农业研究有参考价值。",[364],{"name":359,"url":356},[27,100,28,52,366],"农户认知",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":370},{"candidates":371,"agricultural_evidence":48},[372,376,380],{"code":33,"score":55,"supported":48,"title":373,"abstract":374,"tags":375},[],[28,57,109],[100,28],{"code":197,"score":80,"supported":48,"title":377,"abstract":378,"tags":379},[],[215,198],[],{"code":46,"score":381,"supported":14,"title":382,"abstract":383,"tags":384},2,[],[],[27,52],[386,387],"美国农业 AI ML 调查","农业人工智能 应用障碍","美国农业AIML调查-5306","10.1007\u002Fs11119-026-10459-x",{"doi":389,"openalex_id":391,"authors":392,"venue":359,"cited_by_count":21,"oa_url":356,"card":422,"direction":167,"ingested_from":169},"W7222866367",[393,396,399,402,405,408,411,414,416,419],{"name":394,"orcid":395},"Jennifer Koch","https:\u002F\u002Forcid.org\u002F0000-0002-7067-2705",{"name":397,"orcid":398},"Andrew Fox","https:\u002F\u002Forcid.org\u002F0000-0001-6719-8902",{"name":400,"orcid":401},"Richa Bhattarai","https:\u002F\u002Forcid.org\u002F0000-0003-3827-7761",{"name":403,"orcid":404},"Hank C. Jenkins‐Smith","https:\u002F\u002Forcid.org\u002F0000-0002-5204-2726",{"name":406,"orcid":407},"Jonathan McFadden","https:\u002F\u002Forcid.org\u002F0000-0003-3762-1428",{"name":409,"orcid":410},"Audrey Reinert","https:\u002F\u002Forcid.org\u002F0000-0001-8951-702X",{"name":412,"orcid":413},"Juan Sesmero","https:\u002F\u002Forcid.org\u002F0000-0002-2669-4536",{"name":415,"orcid":8},"M. Z. Walkup",{"name":417,"orcid":418},"Wolfgang Jentner","https:\u002F\u002Forcid.org\u002F0000-0003-1045-6020",{"name":420,"orcid":421},"David S. Ebert","https:\u002F\u002Forcid.org\u002F0000-0001-6177-1296",{"tldr":423,"method":424,"finding":425,"direction":167,"opportunity":426},"通过问卷与访谈调查美国农业利益相关者对AI\u002FML应用的认知与障碍。","在线问卷129人加10人深度访谈的混合方法。","识别出系统之系统、次季节天气预报、精准土壤管理和政策经济四大挑战。","可针对四类障碍开展分区域、分作物类型的采纳意愿追踪与干预研究。","2026-10-19T23:30:02.345893Z",{"id":429,"title":430,"url":431,"summary":432,"summary_zh":433,"content":8,"source_name":434,"source_url":431,"published_at":435,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":436,"sources":439,"tags":441,"primary_topic":444,"secondary_topics":445,"classification":446,"search_phrases":462,"slug":465,"view_count":21,"doi":466,"paper":467,"created_at":481},5192,"A Comparative Study of Client-Level Stabilization Techniques in Federated Learning for Robust Plant Disease Recognition","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fs26206566","The successful integration of artificial intelligence (AI) algorithms in precision agriculture provided farmers with the means to identify potential issues, such as early disease detection or pest infestations, by making crop-related predictions. The new technological developments in this field improve the functioning of traditional agriculture by reducing the dependence on specialized human expertise and facilitating greater access to diagnostic procedures. However, most studies conducted in this area rely on centralized training of AI models, which leads to increased costs and raised privacy-related concerns. Federated learning (FL) offers a compelling framework for collaboratively training machine learning (ML) models on distributed datasets without centralizing the data owned by each participant. FL offers significant advantages in the preservation of data privacy, facilitates the collaborative exploitation of geographically distributed data sources, and reduces the risks associated with centralized collection. In this paper, we investigate the current limitations and performance boundaries of existing Federated and privacy-preserving learning in distributed sensing systems for plant disease detection, in scenarios that closely resemble real-world conditions. Our analysis specifically examined how FL aggregation algorithms, such as FedAvg, FedAvgM and FedProx, work in a custom multi-client experimental setup where data is partitioned so that each participant observes only those symptom patterns associated with specific environmental and climatic conditions. We propose further partial aggregation of the local model, an approach where we send only the model backbone to the central server, while the classification head of each participant remains strictly on the client side. The construction of environmentally heterogeneity is combined with a globally representative stabilizer node designed to preserve knowledge across disease classes that may be absent from certain clients.","人工智能（AI）算法在精准农业中的成功集成，为农民提供了通过作物相关预测来识别潜在问题（如早期病害检测或虫害侵染）的手段。该领域的新技术发展通过减少对专业人类知识的依赖并促进更广泛的诊断程序获取，改善了传统农业的运作方式。然而，该领域的大多数研究依赖于AI模型的集中式训练，这导致成本增加并引发隐私相关担忧。联邦学习（FL）提供了一个引人注目的框架，可在分布式数据集上协作训练机器学习（ML）模型，而无需集中化各参与者拥有的数据。联邦学习在保护数据隐私方面具有显著优势，促进了地理上分布的数据源的协作利用，并降低了与集中收集相关的风险。本文研究了在高度接近真实条件的场景中，现有联邦学习与隐私保护学习在用于植物病害检测的分布式感知系统中的当前局限性与性能边界。我们的分析特别考察了联邦学习聚合算法（如FedAvg、FedAvgM和FedProx）在自定义多客户端实验设置中的工作方式，其中数据被划分，使得每个参与者仅观察到与特定环境和气候条件相关的症状模式。我们提出对本地模型进行进一步的部分聚合，即仅将模型骨干发送至中央服务器，而每个参与者的分类头严格保留在客户端侧。环境异质性的构建与一个全局代表性的稳定器节点相结合，该节点旨在保留某些客户端可能缺失的跨病害类别的知识。","Sensors","2026-10-16T00:00:00Z",{"impact":19,"substance":92,"depth":93,"authority":94,"freshness":437,"relevant":22,"comment":438},9,"该研究针对植物病害识别中的联邦学习客户端稳定化方法提出部分聚合与稳定节点方案，方法新颖、实验贴近真实场景，对智慧农业隐私计算有参考价值，值得进入每日精选。",[440],{"name":434,"url":431},[27,100,31,442,443],"植物病害识别","数据隐私",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":447},{"candidates":448,"agricultural_evidence":48},[449,453,457],{"code":33,"score":47,"supported":48,"title":450,"abstract":451,"tags":452},[],[28,57,109],[100,442,31],{"code":46,"score":113,"supported":48,"title":454,"abstract":455,"tags":456},[],[52,116],[27],{"code":458,"score":22,"supported":14,"title":459,"abstract":460,"tags":461},"I6",[],[],[443],[463,464],"联邦学习 植物病害识别","FedProx 作物病害检测","联邦学习植物病害识别-5192","10.3390\u002Fs26206566",{"doi":466,"openalex_id":468,"authors":469,"venue":434,"cited_by_count":21,"oa_url":431,"card":476,"direction":167,"ingested_from":169},"W7222586734",[470,473],{"name":471,"orcid":472},"Aurora-Felicia Cristea","https:\u002F\u002Forcid.org\u002F0009-0003-9718-065X",{"name":474,"orcid":475},"Ciprian Dobre","https:\u002F\u002Forcid.org\u002F0000-0003-4638-7725",{"tldr":477,"method":478,"finding":479,"direction":167,"opportunity":480},"比较联邦学习客户端级稳定化技术，提升植物病害识别的鲁棒性。","FedAvg\u002FFedAvgM\u002FFedProx聚合，仅上传骨干网络、分类头留本地，","部分聚合加稳定节点可缓解环境异质性与类别缺失带来的性能退化。","可探索面向非独立同分布农业数据的个性化联邦学习与通信效率优化。","2026-10-17T23:30:45.845684Z",{"id":483,"title":484,"url":485,"summary":486,"summary_zh":487,"content":8,"source_name":488,"source_url":485,"published_at":489,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":490,"score_detail":491,"sources":493,"tags":495,"primary_topic":497,"secondary_topics":498,"classification":499,"search_phrases":522,"slug":525,"view_count":21,"doi":526,"paper":527,"created_at":555},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",76,{"impact":181,"substance":92,"depth":93,"authority":94,"freshness":437,"relevant":22,"comment":492},"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。",[494],{"name":488,"url":485},[27,100,29,52,496],"芒果病害",{"code":33,"slug":34,"name":35,"path":36,"parent_code":37},[],{"status":40,"locked":14,"version":41,"method":42,"evidence":500},{"candidates":501,"agricultural_evidence":48},[502,506,510,516],{"code":33,"score":181,"supported":48,"title":503,"abstract":504,"tags":505},[299],[297,299],[100,29],{"code":46,"score":55,"supported":48,"title":507,"abstract":508,"tags":509},[],[27,52,50],[27,52],{"code":511,"score":62,"supported":48,"title":512,"abstract":513,"tags":515},"B2",[],[514],"芒果",[],{"code":517,"score":62,"supported":48,"title":518,"abstract":519,"tags":521},"H8",[],[520],"可持续农业",[],[523,524],"芒果叶片病害 AlexNet","Grad-CAM 芒果病害识别","芒果叶片病害AlexNet-5080","10.1177\u002F1088467x261495708",{"doi":526,"openalex_id":528,"authors":529,"venue":488,"cited_by_count":21,"oa_url":8,"card":550,"direction":262,"ingested_from":169},"W7222328160",[530,533,536,538,541,544,547],{"name":531,"orcid":532},"Anay Ghosh","https:\u002F\u002Forcid.org\u002F0009-0006-6237-7905",{"name":534,"orcid":535},"Tawfikur Rahman","https:\u002F\u002Forcid.org\u002F0000-0002-0456-5143",{"name":537,"orcid":8},"Md Moniruzzaman",{"name":539,"orcid":540},"Nibedita Deb","https:\u002F\u002Forcid.org\u002F0000-0002-1579-1138",{"name":542,"orcid":543},"Hanan Abdullah Mengash","https:\u002F\u002Forcid.org\u002F0000-0002-4103-2434",{"name":545,"orcid":546},"Noorlindawaty Md Jizat","https:\u002F\u002Forcid.org\u002F0000-0001-7618-2283",{"name":548,"orcid":549},"Samir Salem Al‐Bawri","https:\u002F\u002Forcid.org\u002F0000-0003-2852-575X",{"tldr":551,"method":552,"finding":553,"direction":167,"opportunity":554},"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。","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。","2026-10-16T23:30:08.303796Z"]