[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-374":3,"related-374":73},{"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":8,"secondary_topics":32,"classification":33,"search_phrases":68,"slug":71,"view_count":62,"doi":8,"paper":8,"created_at":72},374,"轻量化GSP-YOLO实时杂草识别模型——激光除草场景97.74%精确率、41 FPS推理速度","https:\u002F\u002Fagritechinsights.com\u002Findex.php\u002F2026\u002F08\u002F05\u002Fai-model-gsp-yolo-slashes-weed-battles-without-harming-crops","南京市农业机械化研究所刘赛赛等在Smart Agricultural Technology发表GSP-YOLO模型——基于YOLOv8s的轻量化杂草实时检测系统。该模型将YOLOv8s标准骨干替换为GhostNetV3,并集成空间-通道注意力模块(SCSA),提升对小型、形态相似杂草幼苗的识别能力。在NIAM9weeds数据集上达97.74%精确率,优于YOLOv5n和YOLOv10s等轻量化模型,模型大小减少近30%,计算需求减少超过30%。在边缘设备上通过TensorRT加速,处理速度41 FPS,可支持以步行速度移动的激光除草机器人。温室试验中,系统在0.72 km\u002Fh速度下达93.4%杂草识别准确率,平均275毫秒\u002F株完成除草,作物损伤率低于3.4%。",null,"Researchers in Nanjing have developed a lightweight AI model that could revolutionize how farmers tackle weeds—without harming crops. The breakthrough, published in Smart Agricultural Technology and led by Liu Saisai of the Nanjing Institute of Agricultural Mechanization, introduces GSP-YOLO, a streamlined version of the popular YOLO object detection system, tailored for real-time weed detection in dynamic field conditions.\n\nWeeds may be small, but they’re persistent—and costly. Traditional weeding methods often rely on herbicides or broad-spectrum mechanical removal, which can damage surrounding crops and contribute to environmental strain. Laser weeding, a targeted alternative, requires precise, real-time identification of weed seedlings—especially challenging when weeds look nearly identical to young crops and must be distinguished in seconds.\n\nGSP-YOLO steps in with a targeted solution. By replacing the standard backbone of YOLOv8s with GhostNetV3—a network designed for efficiency—and integrating a new spatial-channel attention module (SCSA), the model sharpens its ability to spot tiny, morphologically similar weed seedlings. “The challenge isn’t just detecting weeds,” says Liu. “It’s doing it fast, accurately, and on a device small enough to ride on a tractor or robot—without draining battery or computing power.”\n\nThe numbers back up the claim. On the NIAM9weeds dataset, GSP-YOLO hit 97.74% precision, outperforming other lightweight models like YOLOv5n and YOLOv10s. Even more striking: it did so while shrinking the model size by nearly 30% and cutting computational demands by over 30%. When deployed on edge devices and accelerated with TensorRT, it processed images at 41 frames per second—fast enough for a laser weeding robot moving at walking speed.\n\nIn live greenhouse trials, the system achieved 93.4% weed recognition accuracy at 0.72 km\u002Fh, removing weeds in an average of 275 milliseconds each, with crop injury kept below 3.4%. That speed and precision could make laser weeding commercially viable for the first time outside small-scale or research settings.\n\nFor farmers, the implications are significant. Laser weeding promises chemical-free, site-specific control that preserves soil health and biodiversity. But until now, the technology has been held back by bulky, slow, or inaccurate detection systems. GSP-YOLO changes that equation. By making AI-powered, real-time weed detection both accurate and deployable on low-power edge devices, it opens the door to scalable, autonomous weeding systems that could operate across thousands of acres.\n\nIndustry observers see this as a turning point. “What we’re seeing here is not just an incremental improvement,” notes one agritech analyst. “It’s a fundamental shift in how we can integrate AI into field operations—lightweight enough for the farm, powerful enough for the task.”\n\nAs robotic platforms become more common in agriculture, models like GSP-YOLO could become the “eyes” of autonomous weeding systems, enabling precise, eco-friendly control in row crops, orchards, and greenhouses. Future developments might see this technology adapted for drone-based scouting or integrated into variable-rate spraying systems, further reducing chemical use.\n\nFor now, Liu and the team are focused on real-world deployment. “Our goal isn’t just to detect weeds,” Liu explains. “It’s to make sure every farmer, regardless of scale, can use this technology to grow more with less—cleaner, smarter, and more sustainably.”\n\nWith precision agriculture evolving rapidly, innovations like GSP-YOLO remind us that sometimes, the smallest models can have the biggest impact.","Smart Agricultural Technology 2026-08-05","2026-08-05T00:00:00Z","论文",10,false,77,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},20,22,18,12,5,1,"轻量化AI模型在激光除草领域取得突破，性能数据详实，对精准农业有实质推动，但时效性稍差。",[25],{"name":10,"url":6},[27,28,29,30,31],"南京农机化所","YOLOv8","激光除草","边缘部署","SmartAgTech",[],{"status":34,"locked":14,"version":35,"method":36,"evidence":37},"needs_review","2026-10-22.2","rules",{"candidates":38,"agricultural_evidence":41,"reason":67},[39,46,52,60],{"code":40,"score":20,"supported":41,"title":42,"abstract":44,"tags":45},"F4",true,[43],"除草",[43],[29],{"code":47,"score":20,"supported":41,"title":48,"abstract":50,"tags":51},"I2",[49],"YOLO",[49],[28],{"code":53,"score":54,"supported":41,"title":55,"abstract":56,"tags":59},"F5",6,[],[57,58],"农业机械","机械化",[],{"code":61,"score":62,"supported":41,"title":63,"abstract":64,"tags":66},"B7",3,[],[65],"温室",[],"多个产业证据相当，需要确认主归属",[69,70],"南京农机化所 激光除草 边缘部署 SmartAgTech","南京农机化所 激光除草","南京农机化所激光除草边缘部署SmartAgTech-374","2026-08-10T23:54:46.220254Z",{"total":54,"page":22,"page_size":54,"items":74},[75,152,216,281,333,393],{"id":76,"title":77,"url":78,"summary":79,"summary_zh":80,"content":8,"source_name":81,"source_url":78,"published_at":82,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":83,"score_detail":84,"sources":91,"tags":93,"primary_topic":8,"secondary_topics":98,"classification":99,"search_phrases":112,"slug":115,"view_count":22,"doi":116,"paper":117,"created_at":151},5136,"Enhanced YOLOv8n for adhered leaf cluster detection in RGB-based fresh green tea leaf sorting","https:\u002F\u002Fdoi.org\u002F10.1016\u002Fj.atech.2026.102640","High-precision detection of adhered leaf clusters (ALCs) based on RGB imaging has long posed a critical challenge in the post-harvest sorting of fresh green tea leaves. To address this issue, this study designed an enhanced YOLOv8n model, named SDSI-YOLO, by integrating the StarNet backbone, DySample upsampling method, and Shape-Inner-IoU loss function to improve ALC detection performance. Experimental results demonstrated that SDSI-YOLO achieved an optimal balance between accuracy and efficiency, with a mean average precision of 92.6% and an F 1 -score of 89.2. It also maintained lightweight characteristics, with a model size of only 3.89 MB and efficient inference speed of 3.1 ms per frame. Ablation experiments confirmed the effectiveness of each module in constructing SDSI-YOLO. The results of visualization tests indicated that the model exhibited a stronger capability in capturing critical details and structural features of ALCs. This research provides a feasible solution for the industrial sorting of fresh green tea leaves, demonstrating considerable application potential.","基于RGB成像的高精度粘连叶簇（adhered leaf clusters，ALCs）检测一直是新鲜绿茶鲜叶采后分选中的关键难题。为解决该问题，本研究通过融合StarNet主干网络、DySample上采样方法和Shape-Inner-IoU损失函数，设计了一种增强型YOLOv8n模型，命名为SDSI-YOLO，以提升ALCs检测性能。实验结果表明，SDSI-YOLO在精度与效率之间实现了最优平衡，平均精度均值达到92.6%，F1分数为89.2。同时，该模型保持了轻量化特性，模型大小仅为3.89 MB，推理速度高达每帧3.1 ms。消融实验验证了各模块在构建SDSI-YOLO中的有效性。可视化测试结果表明，该模型在捕捉ALCs关键细节和结构特征方面表现出更强的能力。本研究为新鲜绿茶鲜叶的工业化分选提供了一种可行方案，具有相当大的应用潜力。","Smart Agricultural Technology","2026-10-14T00:00:00Z",76,{"impact":85,"substance":86,"depth":87,"authority":88,"freshness":89,"relevant":22,"comment":90},16,21,17,13,9,"提出轻量化改进YOLOv8n模型用于鲜茶叶粘连叶簇检测，指标与推理速度均有实测数据，方法新颖且面向产业分选场景，具备精选价值。",[92],{"name":81,"url":78},[94,95,96,28,97],"智慧农业","农业人工智能","目标检测","茶叶分选",[],{"status":34,"locked":14,"version":35,"method":36,"evidence":100},{"candidates":101,"agricultural_evidence":14,"reason":111},[102,106],{"code":47,"score":54,"supported":41,"title":103,"abstract":104,"tags":105},[],[49],[28,95,96],{"code":107,"score":22,"supported":14,"title":108,"abstract":109,"tags":110},"I1",[],[],[94],"农业相关性需要复核",[113,114],"SDSI-YOLO 茶叶分选","鲜叶 粘连叶簇 检测","SDSI-YOLO茶叶分选-5136","10.1016\u002Fj.atech.2026.102640",{"doi":116,"openalex_id":118,"authors":119,"venue":81,"cited_by_count":143,"oa_url":78,"card":144,"direction":148,"ingested_from":150},"W7221817030",[120,122,124,126,128,131,133,136,138,140],{"name":121,"orcid":8},"Jian Wang",{"name":123,"orcid":8},"Yi-jian Liu",{"name":125,"orcid":8},"Yin-yin Lin",{"name":127,"orcid":8},"Cheng-chen Zhang",{"name":129,"orcid":130},"Shengnan Liu","https:\u002F\u002Forcid.org\u002F0000-0001-9886-068X",{"name":132,"orcid":8},"Bin Ji",{"name":134,"orcid":135},"Qiu Wang","https:\u002F\u002Forcid.org\u002F0000-0002-5861-4115",{"name":137,"orcid":8},"Li-hong Du",{"name":139,"orcid":8},"Lei Zheng",{"name":141,"orcid":142},"Fei Ma","https:\u002F\u002Forcid.org\u002F0000-0002-7313-1787",0,{"tldr":145,"method":146,"finding":147,"direction":148,"opportunity":149},"提出改进YOLOv8n模型SDSI-YOLO，实现鲜茶叶粘连叶簇的高精度检测。","融合StarNet主干、DySample上采样与Shape-Inner-IoU损","mAP达92.6%、F1为89.2，模型仅3.89MB、单帧3.1ms，精度与效率兼顾。","农业人工智能与决策模型","可探索轻量检测模型在茶叶分选产线的嵌入式部署与多品类叶簇泛化能力。","openalex","2026-10-17T23:30:03.404999Z",{"id":153,"title":154,"url":155,"summary":156,"summary_zh":8,"content":157,"source_name":158,"source_url":8,"published_at":159,"category":160,"cover_url":8,"hotness":13,"is_selected":14,"score":161,"score_detail":162,"sources":165,"tags":167,"primary_topic":170,"secondary_topics":176,"classification":177,"search_phrases":211,"slug":214,"view_count":143,"doi":8,"paper":8,"created_at":215},4270,"从丰收节看见安徽三农新脉动 中联重科智慧农业服务千万亩次","https:\u002F\u002Fwww.cnr.cn\u002Fah\u002Fahtt\u002F20261005\u002Ft20261005_527834335.shtml","2026年中国农民丰收节安徽活动上，线扫式激光除草机器人效率≥10000株\u002Fh、杂草识别准确率≥90%、作物误杀率≤5%。中联重科智慧农业社会化服务模式以天空地人机一体化感知网络为基础，构建感知-决策-执行-托管全链路闭环，覆盖14个省份累计服务面积超千万亩次，可实现节水30%减肥15%亩均节约128元整体增产10%至15%。","ɨʽݻ򰲻յͼɫзġ˹ݳɱ߸ߡݼԺ̬ѹͻɫֲȱٸЧɫũҵѧе복ѧԺǽܣũͰܾװ޹˾2025зݻˣи߾AIӾ֪άģ뼶ʶӲܹͨ켣滮Ƽ׼ݡЧʡ10000\u002FhӲʶ׼ȷʡ90%ɱʡ5%ݳɹʡ90%װƻͶӦá\n\nԿƼػɫũũˡҲڸ¡ͳ˹Ѳ﷽ʽ£ѲĶҪһСʱ񣬴ض˻߿ɷ120׸߿գ20ӼȫѲũֻװֵ̬ʵʱͣũ°һ鿴һ\n\n˽⣬ؿǻũҵữģʽԡյ˻һ廯֪Ϊũҵģͣ𡰸֪ߡִСйܡȫ·ջǰաϡ½14ʡݣۼƷǧĶΡģʽʵֽˮ30%15%ĶԼͶƷ˹ɱ128Ԫ10%15%\n\n˻Ϸɡڵ⡢AIƶ㡢ũڵܡֳһȽũװǻũҵȰũҵƼ³ɹϤࡣΪȫ13ʳ5ʳʡ֮һռ֡ڵءڼȲӽֵĺơ2025꣬ũۺϻеʴ87%߳ȫ10ٷֵϣũҵƼʴ67.5%߳ȫƽˮƽ3ٷֵϡ","央广网 2026-10-05","2026-10-05T00:00:00Z","报道",66,{"impact":19,"substance":163,"depth":20,"authority":88,"freshness":89,"relevant":22,"comment":164},14,"央媒报道安徽丰收节上中联重科智慧农业社会化服务与激光除草机器人进展，有具体数据但整体偏企业宣传，可作行业动态收录。",[166],{"name":158,"url":155},[94,95,168,29,169],"农业机器人","农业社会化服务",{"code":171,"slug":172,"name":173,"path":174,"parent_code":175},"J4","seasonal-agriculture","农时与节庆活动","\u002Ftopic\u002Fpolicy\u002Fseasonal-agriculture","J",[],{"status":178,"locked":14,"version":35,"method":36,"evidence":179},"classified",{"candidates":180,"agricultural_evidence":41},[181,185,191,196,202,207],{"code":107,"score":20,"supported":41,"title":182,"abstract":183,"tags":184},[94],[94],[94],{"code":171,"score":186,"supported":41,"title":187,"abstract":189,"tags":190},11,[188],"丰收节",[188],[],{"code":40,"score":192,"supported":41,"title":193,"abstract":194,"tags":195},4,[],[43],[29],{"code":197,"score":192,"supported":41,"title":198,"abstract":199,"tags":201},"G2",[],[200],"社会化服务",[169],{"code":203,"score":22,"supported":14,"title":204,"abstract":205,"tags":206},"F6",[],[],[168],{"code":47,"score":22,"supported":14,"title":208,"abstract":209,"tags":210},[],[],[95],[212,213],"中联重科 智慧农业 安徽","激光除草机器人 安徽","中联重科智慧农业安徽-4270","2026-10-06T00:05:02.650986Z",{"id":217,"title":218,"url":219,"summary":220,"summary_zh":8,"content":221,"source_name":222,"source_url":8,"published_at":223,"category":160,"cover_url":8,"hotness":13,"is_selected":14,"score":224,"score_detail":225,"sources":229,"tags":231,"primary_topic":234,"secondary_topics":240,"classification":250,"search_phrases":275,"slug":278,"view_count":279,"doi":8,"paper":8,"created_at":280},2975,"新华社：智能农机打开中国田野新可能——吉林玉米地触摸屏收割机与激光除草机器人","http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316918\u002Fsmart-gears-open-new-possibilities-in-china-fields","新华社9-19报道吉林松源玉米收获机电子控制系统让驾驶员通过单一屏幕管理收获全过程，本土企业年销量已超2000台。第四代激光除草机器人在公主岭试验田作业，由长春理工大学联合产业伙伴研发，作物识别率超99.5%、每小时可清除5-10万株杂草，2026年底预计进入量产。\"十四五\"期间中国农作物耕种收综合机械化率达76.7%、农业无人机超30万架。","[![Image 1: China News](https:\u002F\u002Fassets.kreatio.net\u002Fnn_logos\u002Fchina-news.png)](http:\u002F\u002Fwww.chinanews.net\u002F)\n\n[![Image 2](https:\u002F\u002Fassets.kreatio.net\u002Fnn_logos\u002Fchina-news.png)](http:\u002F\u002Fwww.chinanews.net\u002F)\n\n*   [Home](http:\u002F\u002Fwww.chinanews.net\u002F)\n*   [Asia Pacific News](http:\u002F\u002Fwww.chinanews.net\u002Fcategory\u002Fasia-pacific-news)\n*   [International News](http:\u002F\u002Fwww.chinanews.net\u002Fcategory\u002Fbreaking-international-news)\n*   [Business News](http:\u002F\u002Fwww.chinanews.net\u002Fcategory\u002Fbreaking-business-news)\n*   [Search News](http:\u002F\u002Fwww.chinanews.net\u002Farchive)\n*   [Network](http:\u002F\u002Fwww.themainstreammedia.com\u002F)\n\n*   [![Image 3: facebook](https:\u002F\u002Fassets.kreatio.net\u002Fweb\u002Fnewsnet\u002Fimages\u002Ffacebook.jpg)](http:\u002F\u002Fwww.facebook.com\u002Fpages\u002FChina-News\u002F114750261922882)\n*   [![Image 4: google](https:\u002F\u002Fassets.kreatio.net\u002Fweb\u002Fnewsnet\u002Fimages\u002Fgoogle_pls.jpg)](https:\u002F\u002Fplus.google.com\u002F109476659167282702082)\n*   [![Image 5: rss](https:\u002F\u002Fassets.kreatio.net\u002Fweb\u002Fnewsnet\u002Fimages\u002Frss_feed.jpg)](https:\u002F\u002Ffeeds.chinanews.net\u002Frss\u002F9366300fc9319e9b)\n*   [![Image 6: twitter](https:\u002F\u002Fassets.kreatio.net\u002Fweb\u002Fnewsnet\u002Fimages\u002Ftwitter.jpg)](https:\u002F\u002Ftwitter.com\u002FChinaNewsNet)\n\nSun, 20 Sept 2026\n\n#### [_«_ Back to Home](http:\u002F\u002Fwww.chinanews.net\u002F)\n\n![Image 7: Smart gears open new possibilities in China's fields](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002Fdb2bc74\u002Faa68d274-a88a-4a03-a6dd-1bf1a811f75a\u002F0\u002FXxjwshE000026-20260919-CBMFN0A001-1156x770.jpg)\n\n## [Smart gears open new possibilities in China's fields](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316918\u002Fsmart-gears-open-new-possibilities-in-china-fields#)\n\nXinhua   \n19 Sep 2026, 14:15 GMT+\n\nCHANGCHUN, Sept. 19 (Xinhua) -- As cornfields in northeast China's Jilin Province turn golden for the autumn harvest, bringing in the crop now requires just a few taps on a touchscreen.\n\nInside a corn harvester at Mushen Agricultural Machinery Co., Ltd. in Songyuan, a major corn-producing area, an electronic control system allows the operator to manage every stage of harvesting through a single screen.\n\n\"Ten years ago, we could sell only a few hundred machines a year. Now, our annual sales have reached more than 2,000 at their peak,\" said Wang Xincheng, general manager of the company.\n\nOver the past decade, the share of domestically produced components used in the company's harvesters has risen by about 15 percentage points to 95 percent, while advances in core components and electronic controls have helped narrow the technological gap with comparable foreign products, Wang said.\n\nSome 200 kilometers away in Gongzhuling, another machine is tackling a different field task -- without the use of chemical herbicides.\n\nAt a test field in Changchun's National Agricultural High-Tech Industry Demonstration Zone, a silver-gray, fully electric self-propelled robot glides between rows of corn. An onboard camera scans the field, an AI system distinguishes weeds from crops, and millimeter-level laser beams remove weeds without touching the corn plants.\n\nDeveloped by Changchun University of Science and Technology and its industry partners, the fourth-generation robot has a crop recognition rate of more than 99.5 percent. In field trials, it could remove 50,000 to 100,000 weeds in an hour.\n\nBy using lasers instead of herbicides, the machine provides a solution to reduce chemical usage and helps protect the fertile black soil of northeast China.\n\n\"We designed the robot around the needs of Chinese farming,\" said Wang Xiantao, a member of the research team. \"Each laser-weeding unit works like a building block, allowing the machine to be adapted for various farming scenarios.\"\n\nHe said the robot is expected to enter mass production by the end of 2026.\n\nThe changes seen in Jilin are also unfolding elsewhere in China. From touchscreen harvesters to laser-weeding robots, machines are taking on more farm work and performing it with greater precision.\n\nDuring the 14th Five-Year Plan period (2021-2025), China's comprehensive mechanization rate for crop plowing, sowing and harvesting reached 76.7 percent, with major grain crops basically achieving full-process mechanization, according to the Ministry of Agriculture and Rural Affairs.\n\nMeanwhile, more than 300,000 agricultural drones are now in use nationwide, the largest number in the world.\n\nChina's 15th Five-Year Plan (2026-2030) calls for high-quality development of agricultural machinery and closer integration of farmland, crop varieties, machinery and farming practices. A specific national plan for crop farming development released recently also calls for the development and application of high-end intelligent machinery for large-scale farmland and lighter, more efficient equipment for hilly and mountainous areas.\n\n\"Rapid advances in information technology and AI are becoming deeply integrated with agricultural production,\" said Xia Xianfei, a researcher with the Nanjing Institute of Agricultural Mechanization under the Ministry of Agriculture and Rural Affairs. Smart equipment such as drones and agricultural robots can take over labor-intensive and repetitive tasks while improving farming efficiency, Xia added.\n\nIn Jilin, where the comprehensive mechanization rate for the plowing, sowing and harvesting of major crops has reached 95 percent, research institutes, universities and manufacturers are being brought together to develop smarter agricultural equipment.\n\nIn Henan, another major corn-producing province, BeiDou navigation, 5G communications, new energy technologies and autonomous driving are being integrated into farm machinery to refine the precision and automation of field operations.\n\nIn Xinjiang, research projects launched this year are targeting unmanned cotton-field operations, intelligent sowing, precise control of water, fertilizer and pesticides, and improvements to domestically produced cotton pickers.\n\nChinese-made agricultural machinery is also reaching overseas markets.\n\nChinese cotton pickers are gaining ground in Central Asia, with manufacturers adapting their machines to local planting patterns and harvesting conditions.\n\nThe Xinjiang branch of China Railway Construction Heavy Industry Co., Ltd. delivered a new batch of cotton pickers to Central Asia in July. Orders from the region account for more than 70 percent of its production capacity this year, according to the company.\n\nFeng Guang, vice president of Xinjiang Boshiran Intelligent Agricultural Machinery Co., Ltd., said Chinese high-end cotton pickers can match leading international products in harvesting efficiency while costing about half as much as comparable European and U.S. machines.\n\nWhat is happening in Jilin, Henan, Xinjiang and many other areas reflects a broad shift in how farming is done -- toward greater precision, automation and intelligence.\n\n### [Share article:](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316918\u002Fsmart-gears-open-new-possibilities-in-china-fields#)\n\n Shares \n\n![Image 8: facebook sharing button](https:\u002F\u002Fplatform-cdn.sharethis.com\u002Fimg\u002Ffacebook.svg)Share\n\n![Image 9: twitter sharing button](https:\u002F\u002Fplatform-cdn.sharethis.com\u002Fimg\u002Ftwitter.svg)Post\n\n![Image 10: reddit sharing button](https:\u002F\u002Fplatform-cdn.sharethis.com\u002Fimg\u002Freddit.svg)Share\n\n![Image 11: flipboard sharing button](https:\u002F\u002Fplatform-cdn.sharethis.com\u002Fimg\u002Fflipboard.svg)Flip\n\n![Image 12: email sharing button](https:\u002F\u002Fplatform-cdn.sharethis.com\u002Fimg\u002Femail.svg)Email\n\n[](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279016473\u002Fwhy-becoming-chinese-trend-is-filling-shopping-carts-worldwide)[Why BEIJING, April 28 (Xinhua) -- I'm gonna let you in on a little secret, OK? Tomorrow you are turning Chinese. That's how Sherry Zhu, a Chinese American creator...chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279016473\u002Fwhy-becoming-chinese-trend-is-filling-shopping-carts-worldwide \"Why\")\n\nUndo\n\n[](https:\u002F\u002Fwww.blinkist.com\u002Fmagazine\u002Fposts\u002Fi-asked-chatgpt-what-separates-the-rich-from-everyone-else-tts-chatgpt)[I Asked AI What Rich People Do Differently. The Answer Was One Word.Not stocks. Not crypto. Not connections. The AI kept circling back to something far more ordinary — and far more powerful.Blinkist Magazine | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%202:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%202:)](https:\u002F\u002Fwww.blinkist.com\u002Fmagazine\u002Fposts\u002Fi-asked-chatgpt-what-separates-the-rich-from-everyone-else-tts-chatgpt \"I Asked AI What Rich People Do Differently. The Answer Was One Word.\")\n\nUndo\n\n[](https:\u002F\u002Fwww.blinkist.com\u002Fmagazine\u002Fposts\u002Fthe-5-hour-rule-the-quiet-habit-that-shaped-a-trillionaires-mind)[The Richest Person Alive Spends Five Hours a Week on Something Most People Quit in School It is not networking. It is not investing. It is the one habit that quietly separates the curious from everyone else.Blinkist Magazine | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%203:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%203:)](https:\u002F\u002Fwww.blinkist.com\u002Fmagazine\u002Fposts\u002Fthe-5-hour-rule-the-quiet-habit-that-shaped-a-trillionaires-mind \"The Richest Person Alive Spends Five Hours a Week on Something Most People Quit in School\")\n\nUndo\n\n[](https:\u002F\u002Fmozarreraptions.com\u002F31b1f99e-bb12-4021-b4a6-40f82d1cbb8f)[Wrinkles? Most People Use Lotions. Koreans Do This Instead The Skincare Magazine | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01-a:Below%20Article%20Thumbnails%20|%20Card%204:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01-a:Below%20Article%20Thumbnails%20|%20Card%204:) Read Now](https:\u002F\u002Fmozarreraptions.com\u002F31b1f99e-bb12-4021-b4a6-40f82d1cbb8f \"Wrinkles? Most People Use Lotions. Koreans Do This Instead\")\n\nUndo\n\n[](https:\u002F\u002Fmozarreraptions.com\u002F1150568c-fdba-4b7e-bdca-d481ac558d31)[South Carolina: 52-Year-Old Replaced $21,500 Treatment With This $38 Item\"It Tightens Wrinkles In No Time!\"Anti-Aging | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01-a:Below%20Article%20Thumbnails%20|%20Card%204:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01-a:Below%20Article%20Thumbnails%20|%20Card%204:) Read Now](https:\u002F\u002Fmozarreraptions.com\u002F1150568c-fdba-4b7e-bdca-d481ac558d31 \"South Carolina: 52-Year-Old Replaced $21,500 Treatment With This $38 Item\")\n\nUndo\n\n[](https:\u002F\u002Fwww.thedoctorschannel.com\u002Farticle\u002Fdeep-learning-assisted-analysis-of-biomarker-changes-in-therapy-resistant-neovascular-age-related-macular-degeneration)[Deep Learning-Assisted Analysis of Biomarker Changes in Therapy-Resistant Neovascular Age-Related Macular Degeneration – The Doctor's Channel The Doctor's Channel | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%206:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%206:)](https:\u002F\u002Fwww.thedoctorschannel.com\u002Farticle\u002Fdeep-learning-assisted-analysis-of-biomarker-changes-in-therapy-resistant-neovascular-age-related-macular-degeneration \"Deep Learning-Assisted Analysis of Biomarker Changes in Therapy-Resistant Neovascular Age-Related Macular Degeneration – The Doctor's Channel\")\n\nUndo\n\n[](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278049057\u002Fus-fbi-finds-thousands-of-new-files-on-jfk-assassination)[Trump’s order reveals thousands of hidden JFK assassination documents chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278049057\u002Fus-fbi-finds-thousands-of-new-files-on-jfk-assassination \"Trump’s order reveals thousands of hidden JFK assassination documents\")\n\nUndo\n\n[](https:\u002F\u002Fanti-aging.site\u002Ftbl\u002Fadv-01)[Dermatologists: The 1 Vitamin You Need If You Have Crepey Skin Anti-Aging Center | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=rec-reel-sc2:Below%20Article%20Thumbnails%20|%20Card%207:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=rec-reel-sc2:Below%20Article%20Thumbnails%20|%20Card%207:)](https:\u002F\u002Fanti-aging.site\u002Ftbl\u002Fadv-01 \"Dermatologists: The 1 Vitamin You Need If You Have Crepey Skin\")\n\nUndo\n\n[](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278259626\u002Ffive-nations-elected-to-un-security-council-latvia-joins-first-time)[Five nations elected to UN Security Council, Latvia joins first time chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278259626\u002Ffive-nations-elected-to-un-security-council-latvia-joins-first-time \"Five nations elected to UN Security Council, Latvia joins first time\")\n\nUndo\n\n[](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F272958069\u002Fchinese-veteran-dota-player-somnus-announces-retirement)[Chinese veteran Dota player Somnus announces retirement chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F272958069\u002Fchinese-veteran-dota-player-somnus-announces-retirement \"Chinese veteran Dota player Somnus announces retirement\")\n\nUndo\n\n[](https:\u002F\u002Fwww.thedoctorschannel.com\u002Farticle\u002Fphoton-intravitreal-aflibercept-8-mg-in-diabetic-macular-edema)[PHOTON: Intravitreal Aflibercept 8 mg in Diabetic Macular Edema – The Doctor's Channel The Doctor's Channel | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%208:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%208:)](https:\u002F\u002Fwww.thedoctorschannel.com\u002Farticle\u002Fphoton-intravitreal-aflibercept-8-mg-in-diabetic-macular-edema \"PHOTON: Intravitreal Aflibercept 8 mg in Diabetic Macular Edema – The Doctor's Channel\")\n\nUndo\n\n[MOST POPULAR](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F273852716\u002Ffrench-photographer-captures-real-china-in-hainan)[French photographer captures real China in Hainan chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F273852716\u002Ffrench-photographer-captures-real-china-in-hainan \"French photographer captures real China in Hainan\")\n\nUndo\n\n[MOST POPULAR](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278764195\u002Fhuman-rights-watch-report-highlights-rise-in-sexual-harassment-of-exiled-chinese-activists)[Human Rights Watch report highlights rise in sexual harassment of exiled Chinese activists chinanews](https:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F278764195\u002Fhuman-rights-watch-report-highlights-rise-in-sexual-harassment-of-exiled-chinese-activists \"Human Rights Watch report highlights rise in sexual harassment of exiled Chinese activists\")\n\nUndo\n\n[](https:\u002F\u002Fmozarreraptions.com\u002F30fccaf0-45cd-42af-bfc5-0cfd59078ce3)[52-Year-Old Replaced $21,500 Treatment With This $38 Item Anti-Aging | [Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%2010:)[Sponsored](http:\u002F\u002Fpopup.taboola.com\u002Fen\u002F?template=colorbox&utm_source=newsnet-chinanews&utm_medium=referral&utm_content=thumbs-feed-01:Below%20Article%20Thumbnails%20|%20Card%2010:) Read Now](https:\u002F\u002Fmozarreraptions.com\u002F30fccaf0-45cd-42af-bfc5-0cfd59078ce3 \"52-Year-Old Replaced $21,500 Treatment With This $38 Item\")\n\nUndo\n\n### More China News\n\n[Access More](http:\u002F\u002Fwww.chinanews.net\u002Fcategory\u002Fchina-news)\n\n[![Image 13: EU targets addictive online features for children under 13](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F45023f2\u002F65700c05-f2db-4be6-97e7-76eaf61b3445\u002F1\u002F5-Social-Media-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279313397\u002Feu-targets-addictive-online-features-for-children-under-13)\n\n##### [EU targets addictive online features for children under 13](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279313397\u002Feu-targets-addictive-online-features-for-children-under-13)\n\nChina News.Net\n\n[![Image 14: Yemeni rebels tighten control of Bab el-Mandeb](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002Fffe30b8\u002F213698cf-7d44-4b3d-992d-a93d98783479\u002F2\u002F5-Yemen-Houthis-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279305961\u002Fyemeni-rebels-tighten-control-of-bab-el-mandeb)\n\n##### [Yemeni rebels tighten control of Bab el-Mandeb](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279305961\u002Fyemeni-rebels-tighten-control-of-bab-el-mandeb)\n\nChina News.Net\n\n[![Image 15](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F46db7b7\u002Fff47c73f-61c3-4a9a-9e7b-3f5f9ad38b30\u002F1\u002FCanada-summit-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279305952\u002Fcarney-seeks-global-investment-as-us-trade-war-weighs-on-canada)\n\n##### [Carney seeks global investment as US trade war weighs on Canada](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279305952\u002Fcarney-seeks-global-investment-as-us-trade-war-weighs-on-canada)\n\nChina News.Net\n\n[![Image 16](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002Fc35e3ff\u002Febb345a5-8e55-4e96-a5f7-2e873178a5c1\u002F1\u002FDario-Amodei-2-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279303384\u002Fanthropic-s-amodei-urges-slower-ai-progress-over-safety-risks)\n\n##### [Anthropic's Amodei urges slower AI progress over safety risks](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279303384\u002Fanthropic-s-amodei-urges-slower-ai-progress-over-safety-risks)\n\nChina News.Net\n\n[![Image 17](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002Fb20b8d3\u002F87d213c3-a353-417f-871f-613b29606a46\u002F1\u002FClaudia-Sheinbaum-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279301400\u002Fus-mexico-seek-tariff-deal-as-election-deadline-nears)\n\n##### [US, Mexico seek tariff deal as election deadline nears](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279301400\u002Fus-mexico-seek-tariff-deal-as-election-deadline-nears)\n\nChina News.Net\n\n[![Image 18: Smart gears open new possibilities in China's fields](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F15b0405\u002Faa68d274-a88a-4a03-a6dd-1bf1a811f75a\u002F0\u002FXxjwshE000026-20260919-CBMFN0A001-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316918\u002Fsmart-gears-open-new-possibilities-in-china-fields)\n\n##### [Smart gears open new possibilities in China's fields](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316918\u002Fsmart-gears-open-new-possibilities-in-china-fields)\n\nChina News.Net\n\n[![Image 19](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F97093e9\u002F7cc0d547-0ad2-49e6-9b9a-1a1080df1113\u002F0\u002FXxjpbeE000115-20260919-PEPFN0A001-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316903\u002Fkyrgyzstan-bishkek-chinese-vehicle-green-transportation)\n\n##### [KYRGYZSTAN-BISHKEK-CHINESE VEHICLE-GREEN TRANSPORTATION](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316903\u002Fkyrgyzstan-bishkek-chinese-vehicle-green-transportation)\n\nChina News.Net\n\n[![Image 20](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F815352d\u002F195d8015-9ce2-4766-b6df-b44e8a46247a\u002F0\u002FANI-20260919054259-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316831\u002Fa-heartwarming-feeling-manu-bhaker-on-being-chosen-as-india-flag-bearer-for-asian-games-2026-opening-ceremony)\n\n##### [\"A heartwarming feeling\": Manu Bhaker on being chosen as India's flag-bearer for Asian Games 2026 opening ceremony](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316831\u002Fa-heartwarming-feeling-manu-bhaker-on-being-chosen-as-india-flag-bearer-for-asian-games-2026-opening-ceremony)\n\nChina News.Net\n\n[![Image 21](https:\u002F\u002Fchitra-api.kreatio.com\u002Fapi\u002Fv1\u002Fwps\u002F5668807\u002Fce3c81b7-8b66-4dcf-b97f-db40e13252a1\u002F0\u002FXxjidwE000010-20260919-CBMFN0A001-200x200.jpg)](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316676\u002Faustrian-parliament-speaker-calls-for-mutual-learning-fair-cooperation-between-europe-china)\n\n##### [Austrian parliament speaker calls for mutual learning, fair cooperation between Europe, China](http:\u002F\u002Fwww.chinanews.net\u002Fnews\u002F279316676\u002Faustrian-parliament-speaker-calls-for-mutual-learning-fair-cooperation-between","新华社·中国新闻网","2026-09-19T00:00:00Z",84,{"impact":226,"substance":17,"depth":85,"authority":227,"freshness":89,"relevant":22,"comment":228},24,15,"央媒报道国产智能农机与激光除草机器人突破，含国产化率、识别率等硬数据，产业示范价值突出，值得入选每日精选。",[230],{"name":222,"url":219},[94,168,232,29,233],"智能农机","玉米收获",{"code":235,"slug":236,"name":237,"path":238,"parent_code":239},"A3","maize","玉米","\u002Ftopic\u002Ffield-crops\u002Fmaize","A",[241,246],{"code":53,"slug":242,"name":243,"path":244,"parent_code":245},"machinery","农机装备与机械化","\u002Ftopic\u002Finputs-equipment\u002Fmachinery","F",{"code":40,"slug":247,"name":248,"path":249,"parent_code":245},"crop-protection","农药与绿色防控","\u002Ftopic\u002Finputs-equipment\u002Fcrop-protection",{"status":178,"locked":14,"version":35,"method":36,"evidence":251},{"candidates":252,"agricultural_evidence":41},[253,259,263,267,271],{"code":53,"score":17,"supported":41,"title":254,"abstract":257,"tags":258},[255,256],"农机","收割机",[58],[232],{"code":40,"score":20,"supported":41,"title":260,"abstract":261,"tags":262},[43],[43],[29],{"code":235,"score":186,"supported":41,"title":264,"abstract":265,"tags":266},[237],[237],[],{"code":203,"score":22,"supported":14,"title":268,"abstract":269,"tags":270},[],[],[168],{"code":107,"score":22,"supported":14,"title":272,"abstract":273,"tags":274},[],[],[94],[276,277],"吉林 玉米 智能收割机","长春理工大学 激光除草机器人","吉林玉米智能收割机-2975",2,"2026-09-20T00:03:00.504266Z",{"id":282,"title":283,"url":284,"summary":285,"summary_zh":8,"content":286,"source_name":287,"source_url":8,"published_at":288,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":289,"score_detail":290,"sources":292,"tags":294,"primary_topic":299,"secondary_topics":304,"classification":305,"search_phrases":328,"slug":331,"view_count":279,"doi":8,"paper":8,"created_at":332},794,"方向对齐惩罚(DAP):改进YOLOv8边界框回归的轻量化定位损失(原标题:Directional Alignment Penalty: A Lightweight Localization Loss for Improved Bounding Box Regression in YOLOv8)","https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286","精准农业视觉流水线下游任务(形态特征提取、生长监测、数字孪生同步)直接依赖检测框的几何质量,因此精确定位(而非单纯正确分类)成为关键性能因素。本研究提出方向对齐惩罚(DAP)作为辅助定位正则化项,在不修改检测器架构的情况下将独立加权的归一化中心位移项引入边界框回归目标。DAP-YOLOv8使mAP@0.5:0.95从51.69%提升至53.76%、mAP@0.5从80.03%提升至80.36%,在真实世界智能温室采集的牡蛎蘑菇数据集上重复种子实验进一步表明细粒度定位的提升在不同随机初始化下一致。","## 1. Introduction\n\nSmart agriculture applications rely on computer-vision systems as the backbone for continuous visual monitoring, enabling tasks such as yield estimation, disease detection, autonomous harvesting, and environmental control. Object detection is rarely the final goal of these pipelines: the detected bounding boxes are an intermediate representation from which morphological descriptors (object size, count, maturity, and spatial distribution) are automatically derived and fused with other data streams. Thus, the quality of these descriptors is limited by the geometric accuracy of the detected boxes, and accurate localization (rather than just correct classification) becomes the key performance factor.\n\nThe YOLO family of single-stage detectors [[1](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B1-applsci-16-08286)] is widely used in this area because it predicts object locations and class probabilities in a single forward pass, providing real-time inference with competitive accuracy. In particular, YOLOv8 [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] combines a decoupled detection head, a multi-scale feature aggregation neck and a hierarchical backbone, and has proven effective for continuous agricultural monitoring. The localization component of such detectors is driven by a bounding-box regression loss, and the design of this loss has a strong influence on both the convergence speed and the final localization quality.\n\nMost modern regression losses are formulated based on the intersection-over-union (IoU) between the predicted boxes and ground-truth boxes. To improve gradient behavior, IoU-based variants such as GIoU [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], DIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286)] and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] add geometric constraints based on an enclosing area, a center-distance term or an aspect-ratio penalty. However, in DIoU and CIoU, the normalized center-distance term is integrated into a single objective function with an implicit weight of one (or, for the aspect-ratio term of CIoU, with an adaptive trade-off coefficient), so it cannot be tuned, scheduled or monitored independently of the overlap and aspect-ratio terms. This limitation is most pronounced in agricultural scenes, where objects usually have irregular contours, partial occlusion and dense spatial clustering, as exemplified by the detection of oyster mushroom clusters in a cultivation chamber. To this end, we propose a Directional Alignment Penalty (DAP), an additional lightweight regularization term that explicitly encourages the predicted boxes to move toward their corresponding ground-truth boxes. DAP is measured as the Euclidean distance between the predicted and ground-truth box centers, normalized by the diagonal of the smallest box encapsulating both. We show in [Section 3.4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec3dot4-applsci-16-08286) that this quantity is, up to a square root, exactly the center-distance factor already used inside DIoU and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)]. The contribution of DAP is to decouple that factor from the composite loss and re-express it as a separately weighted additive term whose effect on optimization can be tuned and logged independently of the base regression loss the detector already uses.\n\nThis work makes three contributions. We reformulate the normalized center-distance factor used in DIoU\u002FCIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] as an independently weighted additive regularization term that can be incorporated into existing localization losses without modifying the detector architecture. Second, we demonstrate a practical integration of the penalty into the YOLOv8 training and validation pipeline, including the separate logging of the penalty term for interpretability. Third, we evaluate the method on a dedicated oyster mushroom (Pleurotus ostreatus) dataset collected from a real-world smart greenhouse and demonstrate that stacking DAP on top of the existing CIoU-based box loss of YOLOv8 results in an additional measurable improvement in fine-grained localization, most notably a 2.07 percentage-point improvement in mAP@0.5:0.95, while preserving precision, recall, and the computational efficiency of the base detector.\n\nThe remainder of the paper is organized as follows. [Section 2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec2-applsci-16-08286) reviews related work on single-stage detection and bounding-box regression losses. In [Section 3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec3-applsci-16-08286), we describe the dataset, the YOLOv8 detector, the proposed DAP formulation, its formal relation to DIoU and CIoU, its integration into the detector, and the training protocol. The experimental results are presented in [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286). The implications and limitations of the results are discussed in [Section 5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec5-applsci-16-08286). [Section 6](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec6-applsci-16-08286) concludes the paper.\n\n## 2. Related Work\n\n### 2.1. Deep Learning-Based Object Detection\n\nDeep learning has revolutionized object detection by replacing hand-engineered feature extraction with end-to-end trainable convolutional neural networks (CNNs). Modern object detectors are commonly classified into two-stage and one-stage architectures. Two-stage detectors such as Faster R-CNN first generate region proposals and then classify and refine them, resulting in high localization accuracy but at the cost of increased computational complexity [[1](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B1-applsci-16-08286)]. In contrast, one-stage detectors perform object localization and classification in a single step, enabling substantially faster inference and making them suitable for real-time applications [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)].\n\nAmong one-stage detectors, the YOLO (You Only Look Once) family of detectors has become one of the most influential object detection frameworks due to its excellent speed-accuracy trade-off. The original YOLO architecture [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] and its subsequent versions have continuously improved feature extraction, label assignment, multi-scale representation and localization accuracy. The most recent version, YOLOv8, is now one of the most popular object detectors in real-time computer vision applications [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], thanks to its anchor-free detection approach, decoupled detection heads, improved feature fusion, and optimized training processes. However, despite these architectural improvements, the overall detection performance of YOLOv8 continues to depend strongly on the effectiveness of its bounding-box regression loss.\n\n### 2.2. IoU-Based Bounding Box Regression Losses\n\nBounding-box regression is critical to object detection because it directly determines localization accuracy. Early object detectors used coordinate-based regression losses (e.g., L1 loss and Smooth L1 loss), which considered each coordinate independently. These losses are computationally efficient but correlate weakly with the actual localization quality as measured by the Intersection over Union (IoU) metric. To overcome this limitation, Rezatofighi et al. proposed Generalized IoU (GIoU) [[3](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B3-applsci-16-08286)], which extends the conventional IoU metric by accounting for the smallest enclosing box and can be optimized even when the predicted and ground-truth boxes do not overlap. Later, Zheng et al. proposed Distance IoU (DIoU) to speed up convergence and improve localization accuracy by adding the normalized Euclidean distance between the centers of bounding boxes [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286)]. The same authors further proposed Complete IoU (CIoU) [[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] by incorporating aspect-ratio consistency, resulting in faster convergence and better localization performance than previous IoU-based losses.\n\nBased on the success of DIoU and CIoU, many variants have been proposed to further improve bounding-box regression. Alpha-IoU [[6](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B6-applsci-16-08286)] introduces a power transformation of the IoU term to enable tunable optimization of different localization accuracies and improve robustness against noisy annotations and small datasets. MPDIoU [[7](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B7-applsci-16-08286)] replaces the classical center-distance constraint with the minimum point distances between the corresponding corners of the predicted and ground-truth boxes, which is simpler to compute and also considers the overlap, center displacement, and width-height difference [[7](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B7-applsci-16-08286)]. Among these methods, DIoU and CIoU already include a normalized center-distance term, but this term is merged into a single composite objective with an implicit, fixed weight, so its contribution to optimization is not tunable, schedulable, or monitorable independently of the overlap and aspect-ratio components. Alpha-IoU and MPDIoU jointly reweight or restructure the geometric factors, but the center-alignment contribution is still entangled with the rest of the loss. This limitation is particularly relevant for small, densely distributed, or partially occluded objects, for which the relative weight assigned to center alignment during optimization can be as important as the geometric factor itself.\n\nMotivated by this observation, we revisit the normalized center-distance term already present in DIoU and CIoU [[4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B4-applsci-16-08286),[5](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B5-applsci-16-08286)] and reformulate it as the Directional Alignment Penalty (DAP): an externally added, independently weighted, unsquared regularization component that penalizes the normalized displacement between predicted and ground-truth bounding-box centers. The contribution of DAP is not to introduce a new geometric localization signal but to decouple this factor from the composite DIoU\u002FCIoU formulation. This decoupling allows its weight to be controlled, scheduled, and logged independently of the base regression loss while maintaining full compatibility with existing IoU-based regression losses.\n\n### 2.3. Object Detection in Precision Agriculture\n\nPrecision agriculture has become one of the most promising application domains of artificial intelligence by enabling data-driven crop monitoring, disease diagnosis, yield estimation, and autonomous harvesting. The rapid development of computer vision and deep learning has significantly improved the ability of agricultural systems to detect, classify, and monitor agricultural products under real-world environmental conditions. Compared with traditional machine vision methods based on handcrafted features, deep learning-based object detectors provide greater robustness to variations in illumination, object scale, occlusion, and background complexity, making them highly suitable for agricultural applications [[8](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B8-applsci-16-08286),[9](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B9-applsci-16-08286)].\n\nRecent studies have demonstrated the successful application of deep learning-based object detection across numerous agricultural tasks, including fruit counting, crop growth monitoring, weed identification, disease detection, livestock monitoring, and greenhouse automation [[10](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B10-applsci-16-08286),[11](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B11-applsci-16-08286)]. Among the available detection frameworks, the YOLO family has become particularly attractive because of its favorable trade-off between detection accuracy and computational efficiency. Consequently, various YOLO-based architectures have been adopted for detecting apples, tomatoes, strawberries, cucumbers, peppers, grapes, citrus fruits, wheat heads, rice panicles, and other agricultural products under challenging field conditions [[12](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B12-applsci-16-08286),[13](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B13-applsci-16-08286),[14](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B14-applsci-16-08286)]. These applications require detectors capable of maintaining high localization accuracy despite illumination changes, cluttered backgrounds, overlapping objects, and varying viewing angles.\n\nOne of the most challenging agricultural detection problems is mushroom monitoring in greenhouse environments. Unlike many fruits that are relatively well separated, oyster mushrooms typically grow in dense clusters, exhibit irregular shapes, and frequently overlap at different growth stages. In addition, variations in illumination, shadows, and moisture, together with occlusion by neighboring mushrooms, further complicate accurate localization. These characteristics often reduce the effectiveness of conventional object detectors because even small localization errors may lead to inaccurate size estimation, incorrect maturity assessment, or missed detections during automated harvesting [[15](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B15-applsci-16-08286)].\n\nSeveral recent studies have explored the application of deep learning to mushroom detection and yield estimation. YOLO-based detectors have demonstrated promising performance in identifying mushroom caps, estimating maturity stages, and supporting automated harvesting systems. Researchers have also investigated improvements in feature extraction, attention mechanisms, data augmentation, and image preprocessing to increase detection accuracy under greenhouse conditions [[16](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B16-applsci-16-08286),[17](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B17-applsci-16-08286),[18](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B18-applsci-16-08286),[19](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B19-applsci-16-08286)]. Despite these advances, relatively little attention has been devoted to improving the bounding-box regression objective, even though localization quality directly influences downstream measurements such as mushroom diameter estimation, growth analysis, and digital twin-based monitoring systems. Bounding-box localization is particularly important in mushroom cultivation because multiple mushrooms frequently appear in close proximity and have partially overlapping boundaries. Under these conditions, conventional IoU-based regression losses may produce predicted bounding boxes with similar overlap scores despite differing spatial alignments with the ground truth. Such localization inconsistencies can propagate to subsequent processing stages, affecting object tracking, phenotypic measurements, and yield estimation. Therefore, improving the regression objective rather than modifying the detector architecture offers a lightweight yet effective strategy for enhancing localization accuracy while preserving real-time performance.\n\nMotivated by these difficulties, this work aims to enhance the localization capability of YOLOv8 by reinterpreting the normalized center-distance factor in DIoU\u002FCIoU as the Directional Alignment Penalty (DAP), an additional, independently weighted, unsquared regularization term rather than a novel geometric localization signal. Instead of adding network modules or increasing model complexity, DAP augments the regression loss with this decoupled and independently weighted term. Since the proposed method is only active during training and does not modify the inference pipeline, it maintains the computational efficiency of YOLOv8 while delivering more accurate bounding-box localization, as shown by the observed gains in fine-grained localization ([Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286)). This feature makes DAP particularly well-suited for precision agriculture applications where lightweight models deployable in real time on embedded platforms are highly desirable.\n\n## 3. Materials and Methods\n\n### 3.1. Dataset\n\nThe oyster mushroom (Pleurotus ostreatus) image dataset was specifically constructed in a real-world smart mushroom greenhouse equipped with an RGB camera system and IoT-based monitoring infrastructure [[15](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B15-applsci-16-08286)]. Unlike many of the existing mushroom datasets collected under laboratory conditions or with sparse annotations, this dataset was designed to capture mushrooms in production environments, under variations in illumination, growth stage, object density, and environmental conditions. Images were acquired from two consecutive cultivation cycles in a greenhouse in Mersin, Türkiye, between late December 2022 and late April 2023 using two Full-HD (1920 × 1080) IP cameras. Images were taken during the day and at night to capture varied growth stages and lighting conditions. Each image was time-stamped to match environmental sensor readings.\n\nDuring the two cultivation cycles described above, about 1200 raw greenhouse frames were collected. After frames with no visible mushrooms and images of poor visual quality were excluded, 555 representative full-frame images were retained for object-detection experiments. These images were then split into training and validation subsets containing 499 images and 56 images, respectively. Bounding-box annotations were generated for mushroom instances visible in these full-frame images using the open-source LabelMe [[20](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B20-applsci-16-08286)] tool. Each bounding box was then refined to tightly enclose the mushroom while including minimal background. This procedure improved localization quality during supervised training. One annotation in the training set was excluded from the released label set because its class ID (2) did not match the declared two-class (immature\u002Fmature) configuration; this record was removed before training, rather than remapped, because the correct class could not be determined retrospectively.\n\nA separate, substantially larger set of approximately 8282 maturity-labeled samples was also derived from the same underlying image collection by manually cropping individual mushrooms into separate images assigned to two maturity classes, immature (5124) and mature (3158), with a balanced distribution of day (~4000) and night (~3963) instances. These cropped samples support downstream maturity-related analyses (e.g., per-instance maturity classification) and should not be interpreted as the number of full-frame images used to train the YOLO detector reported in this paper; the YOLOv8 and DAP-YOLOv8 localization experiments in [Section 4](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#sec4-applsci-16-08286) used only the 555-image, full-frame detection dataset (499 training\u002F56 validation) described above. Annotations for the full-frame detection dataset were exported in PASCAL VOC (XML), COCO (JSON), and YOLO Darknet (TXT) formats for compatibility with several detection frameworks.\n\n### 3.2. YOLOv8 Detector\n\nYOLOv8 [[2](https:\u002F\u002Fwww.mdpi.com\u002F2076-3417\u002F16\u002F16\u002F8286#B2-applsci-16-08286)] was adopted as the base detector because its single-pass design maintains high localization accuracy while greatly reducing computational cost, thereby meeting the real-time monitoring and digital-t","MDPI Applied Sciences 16(16):8286 2026-08-20","2026-08-20T00:00:00Z",71,{"impact":20,"substance":17,"depth":19,"authority":20,"freshness":89,"relevant":22,"comment":291},"针对农业场景提出改进YOLOv8定位损失的新方法，有具体实验支撑，但属于细分技术进展，影响范围有限。",[293],{"name":287,"url":284},[295,28,296,297,298],"精准农业","边界框回归","方向对齐惩罚","智能温室",{"code":61,"slug":300,"name":301,"path":302,"parent_code":303},"protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture","B",[],{"status":178,"locked":14,"version":35,"method":36,"evidence":306},{"candidates":307,"agricultural_evidence":41},[308,314,318,324],{"code":107,"score":309,"supported":41,"title":310,"abstract":311,"tags":313},7,[],[295,312],"数字孪生",[295],{"code":61,"score":192,"supported":41,"title":315,"abstract":316,"tags":317},[],[65],[298],{"code":319,"score":62,"supported":41,"title":320,"abstract":321,"tags":323},"B5",[],[322],"蘑菇",[],{"code":47,"score":22,"supported":14,"title":325,"abstract":326,"tags":327},[],[],[28],[329,330],"方向对齐惩罚 边界框回归 智能温室 精准农业","方向对齐惩罚 边界框回归","方向对齐惩罚边界框回归智能温室精准农业-794","2026-08-21T00:05:57.684831Z",{"id":334,"title":335,"url":336,"summary":337,"summary_zh":8,"content":338,"source_name":339,"source_url":8,"published_at":340,"category":160,"cover_url":8,"hotness":13,"is_selected":14,"score":341,"score_detail":342,"sources":344,"tags":346,"primary_topic":350,"secondary_topics":355,"classification":361,"search_phrases":388,"slug":391,"view_count":279,"doi":8,"paper":8,"created_at":392},665,"黑龙江天工开悟农业大模型推广至全省114个农场 克豆91高蛋白大豆以3155万元转让","https:\u002F\u002Fwww.hljapp.com\u002Fcontent\u002Fdetail\u002F3928","黑龙江省新质生产力科技成果展在哈尔滨工业大学先进技术研究院开展,71项科创成果集中亮相。哈尔滨华工智耘科技激光除草机器人在黑龙江大豆、玉米、白菜等作物种植基地开展多场景验证,累计应用面积超1000亩,杂草净除率超95%,伤苗率小于0.1%。天工开悟农业大模型将选种、施肥、田间管理、收获等环节纳入数据分析,已推广至全省114个农场。黑龙江省农业科学院克山分院董全中团队历时10年选育的高蛋白大豆新品种克豆91,2025年生产经营权以3155万元转让,刷新国内大豆单品种单次市场化交易纪录。","育秧巡检机器人穿行苗床，“悟空号”无人无缆潜水器探向万米深海，新型材料、数字技术等创新成果依次呈现……近日，黑龙江省新质生产力科技成果展在哈尔滨工业大学先进技术研究院开展，来自全省高校院所、重点企业的71项科创成果集中亮相，一件件“龙江智造”勾勒出科技创新加快转化为现实生产力的清晰轨迹。\n\n 从实验室走向生产线，从技术突破迈向产业应用，在这片黑土地上，创新之“进”，正不断拓展发展之“新”。\n\n 在新时代东北全面振兴的新征程上，黑龙江持续推动科技创新和产业创新深度融合，加快把创新“关键变量”转化为振兴发展“最大增量”。\n\n**向纵深处“进”：传统优势焕发新机**\n\n 夏管时节，黑土地绿意正浓。在黑龙江一处速冻青刀豆种植基地里，四轮激光除草机器人沿垄缓缓前行，多光谱相机和深度传感器持续扫描，算法迅速分辨作物与杂草，激光束随即精准击中杂草生长点。\n\n “过去除草主要靠人工和化学药剂，要么成本高，要么影响作物和土壤。”哈尔滨华工智耘科技有限公司副总经理李涌伟说，激光除草是纯绿色物理除草方式，杂草净除率超过95%，伤苗率小于0.1%，每小时最高可毁伤上百万棵杂草，成本较人工除草最高可下降70%。田间试验显示，激光除草可减少杂草对作物养分争夺，实现作物增产。\n\n 看得准、打得准，一束光成为除草的新工具。目前，该装备已在黑龙江省内大豆、玉米、白菜等作物种植基地开展多场景验证，累计应用面积超过1000亩。核心激光光源、人工智能视觉算法和植物识别模型均由团队自主研发，国产高端激光除草装备正从田间试验走向规模化应用。\n\n 田垄之外，农业大模型正在覆盖更广阔的农场。“天工开悟”农业大模型将选种、施肥、田间管理、收获等环节纳入数据分析，已推广至全省114个农场。气象、土壤、作物长势等信息汇聚起来，让过去更多依靠经验作出的农事判断，开始有了更精准的数据支撑。\n\n 田间管理向“智”而行，种源创新则关乎农业的长远竞争力。黑龙江省农业科学院克山分院董全中团队历时10年选育的高蛋白大豆新品种“克豆91”，2025年生产经营权以3155万元转让，刷新国内大豆单品种单次市场化交易纪录。一个新品种由实验室走向市场，也让科研成果的价值有了可以丈量的产业尺度。\n\n “一台装备改变一种作业方式，一个模型连接多个生产环节，农业科技，就是要让种子更优良，让农事更精准，让粮食更优质。”黑龙江省科学技术厅科技二处处长杨雪说。科技创新加快嵌入耕、种、管、收农业生产全过程，为大面积单产提升注入动力，让“中国饭碗”装进更多优质龙江粮。\n\n 田野之外，科技创新也在重新擦亮龙江制造的招牌。在中国一重生产车间，3000吨级加氢反应器等重大装备接续突破；在哈电集团，超超临界循环流化床锅炉、大型水电机组等不断刷新技术指标。\n\n 曾经以“块头大、分量重”闻名的龙江制造，正把新材料、数字技术和智能制造加快融入设计、加工、检测等环节，推动大国重器向更高端、更智能、更绿色迈进。\n\n 把传统优势转化为发展胜势，新质生产力加快向产业纵深渗透，新技术、新要素正让厚实的龙江家底焕发新的生机。\n\n**向前沿处“进”：新兴赛道塑造新动能**\n\n “老树发新枝”，也要让“新苗长成大树”。当前，黑龙江着眼科技变革和产业演进趋势，以优势资源切入细分赛道，以技术突破牵引产业集聚，新的增长点正由单点萌发走向链式生长。\n\n 向高处攀登，商业航天加速起势。\n\n 走进哈尔滨新区工大卫星智造基地，洁净厂房内，卫星总装、测试和环境试验有序衔接。\n\n 哈尔滨工大卫星技术有限公司成立于2020年。短短数年，企业已完成14次发射任务，将37颗卫星送入太空，涵盖通信、导航、遥感、深空探测等领域，如今，其研发试验总部基地具备年产180颗500公斤量级卫星的能力。\n\n “柔性制造不是把单星研制简单复制，而是通过标准化模块、数字化工艺和流程再造，提高卫星批量生产和稳定交付能力，为低轨星座快速组网提供支撑。”工大卫星总经理助理王群说。\n\n 卫星制造从单星定制走向柔性化、批量化、数字化，带来的不只是产能提升，也在牵动更长的产业链。标准化模块提高生产效率，数字化工艺贯通研制流程，稳定的批量需求则牵引载荷研制、关键零部件、测运控和数据应用等环节加快集聚。以工大卫星为“链主”，黑龙江商业航天产业正由一颗星、一个企业，向一条链、一个产业集群延伸。\n\n 向寒处探索，“冷资源”催生新产业。\n\n 今年2月，在哈尔滨冰雪大世界科技赋能冰雪实验区，一台智能大型冰雕自动化搭建机器人进入测试阶段。自动上料、精准提升、抓取定位、喷淋黏合，重达1吨的冰块被逐层垒起，以毫米级精度完成砌筑。投入应用后，机器人可承担部分高空作业，降低极寒环境下的施工风险。\n\n 与它协同工作的，还有江面采冰、冰块切割、开槽、雕刻和表面打磨等系列装备，过去主要依靠人工和传统机械完成的冰建流程，开始有了成套智能装备的参与，向全流程智能化、标准化迈进。\n\n “冰雪大世界和松花江采冰场就是我们的试验场，低温导航、视觉识别、精密控制等技术攻关在真实环境中反复接受检验、不断迭代。”哈尔滨冰科智能装备有限公司董事长姜文锐说。\n\n 从冰雪消费到冰雪制造，黑龙江漫长的冬季正在释放新的产业价值。文旅端涌动的热度随之向研发制造端传导，“冷资源”不再只是季节性消费场景，也成为新技术、新产品的孵化场。\n\n 向微处进发，机器人技术进入更加精细的产业场景。\n\n 在哈尔滨灵博机器人有限公司的“创造车间”里，几款灵巧手原型机正在反复调试，抓取、提拉、调整姿态，一系列动作流畅完成。\n\n “我们在这样小的空间里集成了15个主动自由度和多类传感器。”哈尔滨灵博机器人有限公司总经理祁乐说，不仅要“抓得住”，还要能够感知接触力度，在与不同物体交互时实时调整动作。\n\n 灵巧背后，是多年技术积累。灵博团队来自哈工大机器人技术与系统全国重点实验室空间机器人团队。面向具身智能兴起的新赛道，曾经服务航天重大工程的灵巧操作、力感知和控制技术，开始进入人形机器人、工业制造等应用场景，打开了一片新的产业空间。\n\n 科教资源孕育技术源头，工业积淀支撑工程转化，特色场景推动产品迭代。黑龙江正在新的产业坐标中重新发现自身优势，并将其转化为新产业、新企业和新动能。\n\n**向产业处“进”：创新成果跑出加速度**\n\n “十五五”规划纲要提出，推动科技创新和产业创新深度融合。对科研资源富集的黑龙江而言，能否让更多成果走出实验室、进入生产线、汇入产业链，是科技优势转化为振兴动能的关键一跃。\n\n 近年来，黑龙江国家级科技创新平台基地由46家增至69家，其中，全国重点实验室由7家增至16家。机器人技术与系统、陆相页岩油等高水平平台相继布局，为基础研究、关键技术攻关和产业升级注入源头活水。\n\n 科技成果从“最初一公里”走到“最后一公里”，是一场基础研究、技术攻关、中试熟化、企业孵化和场景应用的接力。\n\n 针对一些实验室成果“不成熟”、企业“接不住”等难题，黑龙江布局31家概念验证中心，研判技术可行性和市场潜力，帮助成果找准转化方向；建设99家中试熟化平台，着力破解工艺放大、稳定生产和成本控制等工程化难题。目前，相关平台已推动255个项目开展中试熟化，其中135个实现产业化。\n\n 黑龙江推动企业从成果承接者更多走向创新组织者，生产一线出题、高校院所协同解题、市场检验答卷的路径日益清晰。“十四五”期间，全省攻克重点领域关键技术719项，企业牵头项目资金比例由61%提高到71%。\n\n 让企业挑起创新大梁，还需要科研、人才、资本和场景等要素协同支撑。黑龙江围绕大学、大院、大所建设7个创新创业生态圈，集聚科技型企业2512家，推动科研团队、孵化平台、产业资本和应用场景加快对接。实验室里的技术和样机，由此加快完成工程化验证、产品化打磨和市场化检验，进而转化为产品、孵化出企业、带动起产业。\n\n 制度改革也在为成果转化松绑。职务科技成果单列管理、转化类人才职称评价、尽职免责等改革陆续推进，逐步化解科研人员“不敢转、不愿转”的顾虑。2023年以来，黑龙江省累计转化重大科技成果1873项，新增经济效益348.43亿元，成果转化逐步走向有组织、全链条推进。\n\n 创新链上的一棒接着一棒跑，一个个科技创新“点”由此连成产业发展“线”。黑龙江省科学技术厅科技规划处处长孙启华表示，“十五五”期间，黑龙江将做强有组织科研和有组织转化，围绕产业需求优供给，聚焦成果转化提效能，整合创新资源强企业，培育壮大创新人才队伍，加快以科技创新引领产业全面振兴。\n\n 以创新之“进”，促产业之“新”、动能之“新”、发展之“新”。黑土地上，传统产业因科技赋能焕发新机，新兴产业在创新积累中加快起势，更多科技成果从实验室走向生产线、转化为产业竞争力，振兴发展的科技底色愈加鲜明。\n\n（备注：青蛙工业云平台是哈尔滨本地的一家工业互联网平台，致力于助力黑龙江省各地工业企业完成数字化转型，为推动新质生产力的迅猛发展、实现网络协同制造贡献力量，进而帮助企业降本增效，达成共赢局面。\n\n青蛙工业云平台严格把控数字转型方案的质量，结合哈尔滨在数字化转型城市建设、数字化车间打造、软件定制、互联网工业发展、标识解析方案以及数字化转型咨询诊断等方面的产业政策，为黑龙江工业企业的数字化转型升级探索新路径，提供能够快速有效提升产量与增量的解决方案，助力哈尔滨工厂实现转型，推动黑龙江工业振兴。）","黑龙江日报客户端 2026-08-17","2026-08-17T00:00:00Z",75,{"impact":18,"substance":19,"depth":85,"authority":13,"freshness":89,"relevant":22,"comment":343},"报道黑龙江农业科技与产业创新融合，涵盖农业大模型、高蛋白大豆转让等亮点，省级官媒，内容详实，但部分内容偏向综合展示。",[345],{"name":339,"url":336},[347,29,348,349],"黑龙江省农科院","天工开悟","克豆91",{"code":351,"slug":352,"name":353,"path":354,"parent_code":239},"A4","soybean-oilseeds","大豆与油料","\u002Ftopic\u002Ffield-crops\u002Fsoybean-oilseeds",[356],{"code":47,"slug":357,"name":358,"path":359,"parent_code":360},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":178,"locked":14,"version":35,"method":36,"evidence":362},{"candidates":363,"agricultural_evidence":41},[364,369,374,378,382],{"code":351,"score":20,"supported":41,"title":365,"abstract":367,"tags":368},[366],"大豆",[366],[349],{"code":47,"score":186,"supported":41,"title":370,"abstract":372,"tags":373},[371],"大模型",[371],[],{"code":40,"score":192,"supported":41,"title":375,"abstract":376,"tags":377},[],[43],[29],{"code":235,"score":62,"supported":41,"title":379,"abstract":380,"tags":381},[],[237],[],{"code":383,"score":62,"supported":41,"title":384,"abstract":385,"tags":387},"F3",[],[386],"施肥",[],[389,390],"黑龙江省农科院 天工开悟 激光除草 克豆91","黑龙江省农科院 天工开悟","黑龙江省农科院天工开悟激光除草克豆91-665","2026-08-17T23:54:54.777166Z",{"id":394,"title":395,"url":396,"summary":397,"summary_zh":8,"content":398,"source_name":399,"source_url":8,"published_at":400,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":401,"score_detail":402,"sources":404,"tags":406,"primary_topic":410,"secondary_topics":411,"classification":412,"search_phrases":428,"slug":431,"view_count":62,"doi":8,"paper":8,"created_at":432},540,"AdapterMoE: A Two-Stage Hard-Routing Mixture-of-Experts Architecture for Multi-Crop Disease Recognition with Calibrated Rejection and Incremental Learning","https:\u002F\u002Farxiv.org\u002Fabs\u002F2608.08808","【预印本,未经同行评审】针对多作物病害识别任务,提出基于确定性硬路由的两阶段 AdapterMoE 架构:RouterHead 依据最大 softmax 概率阈值分类作物并拒判非目标作物,Energy+KNN 双重门控模块捕获分布外输入;在冻结 EfficientNet-B0 主干上添加五个逐作物 Adapter,通过温度缩放校准。在 PlantVillage(5 作物 26 类)上,AdapterMoE 与最优基线 Macro-F1 差距在 0.24 分以内,但训练成本约为全网络基线的 9%,并可通过 add_crop 接口实现逐作物增量更新,避免专家坍缩,适合田间部署成本敏感的精准植保场景。","[View PDF](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2608.08808)[HTML (experimental)](https:\u002F\u002Farxiv.org\u002Fhtml\u002F2608.08808v1)\n\n> Abstract:Timely crop-disease identification is critical to food security. Multi-crop recognition suits Mixture-of-Experts (MoE), but conventional soft-routing MoE learns crop assignment freely end-to-end, letting a few experts dominate (expert collapse) with no semantic correspondence to crops, and facing high retraining costs, unstable rejection of non-target inputs, and a saturated accuracy ceiling. We shift the objective from accuracy toward a trade-off among deployment cost, scaling flexibility, and rejection stability, using deterministic hard routing. We propose AdapterMoE: a RouterHead classifies the crop and rejects non-target crops via a Maximum Softmax Probability threshold, with a dual-gate Energy+KNN out-of-distribution module catching distribution-shifted inputs; five per-crop Adapters atop a frozen EfficientNet-B0 backbone discriminate diseases, each calibrated via Temperature Scaling. Because experts are hard-isolated at the data level, the design avoids expert collapse and exposes an add_crop interface for local, per-crop updates instead of full retraining. On PlantVillage (5 crops, 26 classes), across a fair five-system comparison, AdapterMoE attains accuracy statistically indistinguishable from the best baselines (Macro-F1 within a 0.24-point band) while cutting training cost to about 9% of full-network baselines, expanding to a new crop in\n\nComments:17 pages, 7 figures, 14 tables\nSubjects:Computer Vision and Pattern Recognition (cs.CV)\nCite as:[arXiv:2608.08808](https:\u002F\u002Farxiv.org\u002Fabs\u002F2608.08808) [cs.CV]\n(or [arXiv:2608.08808v1](https:\u002F\u002Farxiv.org\u002Fabs\u002F2608.08808v1) [cs.CV] for this version)\n[https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2608.08808](https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2608.08808)\n\narXiv-issued DOI via DataCite\n\n## Submission history\n\nFrom: PinHsun Huang [[view email](https:\u002F\u002Farxiv.org\u002Fshow-email\u002F12891f91\u002F2608.08808)] \n\n**[v1]** Sun, 9 Aug 2026 16:45:21 UTC (106 KB)","arXiv [预印本]","2026-08-09T16:45:21Z",78,{"impact":19,"substance":18,"depth":17,"authority":20,"freshness":54,"relevant":22,"comment":403},"提出农业多作物病害识别的硬路由MoE架构，方法新颖且实验扎实，对智慧农业有实质贡献。",[405],{"name":399,"url":396},[95,295,30,407,408,409],"植物病害识别","Mixture-of-Experts","多作物",{"code":40,"slug":247,"name":248,"path":249,"parent_code":245},[],{"status":178,"locked":14,"version":35,"method":36,"evidence":413},{"candidates":414,"agricultural_evidence":41},[415,420,424],{"code":40,"score":62,"supported":41,"title":416,"abstract":417,"tags":419},[],[418],"植保",[],{"code":47,"score":279,"supported":14,"title":421,"abstract":422,"tags":423},[],[],[95,407],{"code":107,"score":22,"supported":14,"title":425,"abstract":426,"tags":427},[],[],[295],[429,430],"农业人工智能 植物病害识别 精准农业 边缘部署","农业人工智能 植物病害识别","农业人工智能植物病害识别精准农业边缘部署-540","2026-08-15T00:02:55.709784Z"]