{"id":88,"date":"2023-01-02T19:59:37","date_gmt":"2023-01-02T11:59:37","guid":{"rendered":"http:\/\/106.52.213.145:21080\/?p=88"},"modified":"2023-02-16T02:54:21","modified_gmt":"2023-02-15T18:54:21","slug":"ydbjgxygtxmbjcsfzs3","status":"publish","type":"post","link":"https:\/\/apifj.com\/index.php\/2023\/01\/02\/ydbjgxygtxmbjcsfzs3\/","title":{"rendered":"[\u9605\u8bfb\u7b14\u8bb0] \u5149\u5b66\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u7b97\u6cd5\u7efc\u8ff0\uff08\u4e09\uff09"},"content":{"rendered":"

\u5149\u5b66\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u7b97\u6cd5\u7efc\u8ff0\uff08\u4e09\uff09<\/h1>\n

\u539f\u6587pdf\u4e0b\u8f7d\uff1a\u4e0b\u8f7d\u94fe\u63a5<\/a><\/p>\n

4. \u6570\u636e\u96c6\u548c\u7b97\u6cd5\u6027\u80fd\u6bd4\u8f83<\/h2>\n

4.1 \u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u6570\u636e\u96c6<\/h3>\n

\u5e38\u89c1\u6570\u636e\u96c6<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
\u8bba\u6587<\/th>\n\u6570\u636e\u96c6\u540d\u79f0<\/th>\n<\/tr>\n<\/thead>\n
[127]Heitz G, Koller D. Learning spatial Context: Using stuff to find things. In: Proceedings of the 10th European Conference on Computer Vision (ECCV 2008). Marseille, France: Springer, 2008. 30\u221243<\/td>\nTAS \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[128]Maggiori E, Tarabalka Y, Charpiat G, Alliez P. Can semantic labeling methods generalize to any city? The inria aerial image labeling benchmark. In: Proceedings of the 2017 IEEE Interna- tional Geoscience and Remote Sensing Symposium (IGARSS). Fort Worth, USA: IEEE, 2017. 3226\u22123229<\/td>\nSZTAKI-INRIA \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[129]Cheng G, Han J W, Zhou P C, Guo L. Multi-class geospatial object detection and geographic image classification based on collection of part detectors. ISPRS Journal of Photogram- metry and Remote Sensing, 2014, 98<\/strong>: 119\u2212132<\/td>\nNWPU VHR-10 \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[130]Razakarivony S, Jurie F. Vehicle detection in aerial imagery: A small target detection benchmark. Journal of Visual Commu- nication and Image Representation, 2016, 34<\/strong>: 187\u2212203<\/td>\nVEDAI \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[131]Zhu H G, Chen X G, Dai W Q, Fu K, Ye Q X, Jiao J B. Ori- entation robust object detection in aerial images using deep convolutional neural network. In: Proceedings of the 2015 IEEE International Conference on Image Processing (ICIP). Quebec City, Canada: IEEE, 2015. 3735\u22123739<\/td>\nUCAS-AOD \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[132]Liu K, Mattyus G. Fast multiclass vehicle detection on aerial images. IEEE Geoscience and Remote Sensing Letters, 2015, 12<\/strong>(9): 1938\u22121942<\/td>\nDLR 3K Vehicle \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[133]Liu Z K, Yuan L, Weng L B, Yang Y P. A high resolution op- tical satellite image dataset for ship recognition and some new baselines. In: Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods (ICPRAM 2017). Porto, Portugal: SciTe Press, 2017. 324\u2212331<\/td>\nHRSC2016 \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[79]Long Y, Gong Y P, Xiao Z F, Liu Q. Accurate object localiza- tion in remote sensing images based on convolutional neural networks. IEEE Transactions on Geoscience and Remote Sens- ing, 2017, 55<\/strong>(5): 2486\u22122498<\/td>\nRSOD \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[119]XiaGS,BaiX,DingJ,ZhuZ,BelongieS,LuoJB,etal. DOTA: A large-scale dataset for object detection in aerial im- ages. In: Proceedings of the 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE, 2018. 3974\u22123983<\/td>\nDOTA \u6570\u636e\u96c6<\/td>\n<\/tr>\n
[4]Li K, Wan G, Cheng G, Meng L Q, Han J W. Object detec- tion in optical remote sensing images: A survey and a new benchmark. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 159<\/strong>: 296\u2212307<\/td>\nDIOR \u6570\u636e\u96c6<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

4.2 \u901a\u7528\u76ee\u6807\u68c0\u6d4b\u7b97\u6cd5\u6027\u80fd\u6bd4\u8f83<\/h3>\n

\u8bc4\u4ef7\u56e0\u7d20<\/h4>\n

\u7cbe\u786e\u5ea6 (Precision)\uff1a\u88ab\u68c0\u6d4b\u51fa\u6765\u7684\u76ee\u6807\u4e2d, \u68c0\u6d4b\u6b63\u786e\u7684\u6982\u7387\u3002<\/p>\n

\u53ec\u56de\u7387 (Recall)\uff1a\u6620\u6240\u6709\u5f85\u68c0\u6d4b\u76ee\u6807\u4e2d\u88ab\u6210\u529f\u68c0\u6d4b\u5230\u7684\u6982\u7387\u3002<\/p>\n

\u7cbe\u5ea6\u2212\u53ec\u56de\u7387\u66f2\u7ebf\uff08PR\u66f2\u7ebf\uff09\uff1a\u4ee5\u53ec\u56de\u7387\u4e3a\u6a2a\u5750\u6807\u3001\u7cbe\u786e\u5ea6\u4e3a\u7eb5\u5750\u6807\u753b\u51fa\u7684\u66f2\u7ebf\u3002<\/p>\n

\u5e73\u5747\u7cbe\u5ea6\uff08AP\uff09\uff1aPR\u66f2\u7ebf\u4e0b\u5bf9\u5e94\u7684\u9762\u79ef\uff0c\u5355\u4e00\u7c7b\u522b\u7684\u68c0\u6d4b\u6027\u80fd\u3002<\/p>\n

\u5e73\u5747 AP \u503c\uff08mAP\uff09\uff1a\u6240\u6709\u7c7b\u522b\u7684\u5e73\u5747 AP \u503c\u3002<\/p>\n

NWPU VHR-10 \uff1a\u7c7b\u522b\u6570\u8f83\u5c11, \u4efb\u52a1\u76f8\u5bf9\u7b80\u5355, \u662f\u65e9\u671f\u6df1\u5ea6\u5b66\u4e60\u7b97\u6cd5\u5e7f\u6cdb\u4f7f\u7528\u7684\u6570\u636e\u96c6;<\/p>\n

DOTA \uff1a\u6570\u636e\u96c6\u66f4\u52a0\u590d\u6742, \u68c0\u6d4b\u96be\u5ea6\u66f4\u5927, \u662f\u73b0\u9636\u6bb5\u7684\u4e3b\u8981\u8bc4\u6d4b\u6570\u636e\u96c6<\/p>\n

\"\"<\/p>\n

\u8868\u683c\u5f97\u5230\u7684\u4fe1\u606f\uff1a<\/p>\n

\uff081\uff09\u5728\u5904\u7406\u66f4\u590d\u6742\u548c\u66f4\u6709\u6311\u6218\u6027\u7684\u95ee\u9898\u65f6\uff0c\u5f88\u591a\u95ee\u9898\u76ee\u524d\u8fd8\u65e0\u6cd5\u5f97\u5230\u6709\u6548\u89e3\u51b3, \u73b0\u6709\u7814\u7a76\u6210\u679c\u8fd8\u65e0 \u6cd5\u6ee1\u8db3\u9ad8\u6807\u51c6\u5e94\u7528\u573a\u5408\u9700\u6c42\u3002<\/p>\n

\uff082\uff09\u9488\u5bf9\u9065\u611f\u56fe\u50cf\u76ee\u6807\u81ea\u8eab\u7279\u70b9\u8fdb\u884c\u7684\u4e00\u7cfb\u5217\u6539\u8fdb\u662f\u6709\u6548\u7684\u3002<\/p>\n

\uff083\uff09\u9274\u4e8e\u65cb\u8f6c\u6846\u5bf9\u76ee\u6807\u7684\u5b9a\u4f4d\u66f4\u52a0\u7cbe\u7ec6\uff0c\u65cb\u8f6c\u6846\u68c0\u6d4b\u5728\u9065\u611f\u9886\u57df\u5e94\u7528\u6f5c\u529b\u4e0d\u65ad\u63d0\u9ad8\uff0c\u65cb\u8f6c\u6846\u68c0\u6d4b\u7684 mAP \u503c\u8981\u7565\u4f4e\u4e8e\u6c34\u5e73\u6846\uff08\u56e0\u4e3a\u66f4\u4e25\u683c\uff09\uff0c\u96be\u5ea6\u66f4\u9ad8\u3002<\/p>\n

\uff084\uff09\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u6027\u80fd\u63d0\u5347\u5f88\u5927\u7a0b\u5ea6\u4f9d\u8d56\u6df1\u5ea6\u5b66\u4e60\u81ea\u8eab\u7684\u53d1\u5c55\u3002<\/p>\n

\uff085\uff09\u56fe\u50cf\u5206\u5757\u7684\u65b9\u5f0f\u5df2\u7ecf\u6210\u4e3a DOTA \u4e0a\u516c\u8ba4\u7684\u6807\u51c6\u9884\u5904\u7406\u65b9\u5f0f\u3002<\/p>\n

\uff086\uff09\u591a\u5c3a\u5ea6\u7279\u5f81\u91d1\u5b57\u5854\u8fdb\u884c\u68c0\u6d4b\u7684\u65b9\u5f0f, \u53ef\u4ee5\u6709\u6548\u5904\u7406\u76ee\u6807\u5c3a\u5ea6\u8303\u56f4\u53d8\u5316\u5927\u7684\u95ee\u9898\u3002<\/p>\n

\uff087\uff09\u73b0\u6709\u9488\u5bf9\u5bc6\u96c6\u5206\u5e03\u3001\u5916\u89c2\u6a21\u7cca\u7684\u76ee\u6807\u68c0\u6d4b\u7814\u7a76\u6210\u679c\u5e76\u4e0d\u591a\u3002<\/p>\n

5. \u73b0\u5b58\u95ee\u9898\u548c\u53d1\u5c55\u8d8b\u52bf<\/h2>\n

5.1 \u73b0\u5b58\u95ee\u9898<\/h3>\n

\uff081\uff09 \u9488\u5bf9\u8d85\u5927\u56fe\u50cf\u5c3a\u5bf8\u7684\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b, \u73b0\u6709\u65b9\u6cd5\u5c1a\u4e0d\u80fd\u76f4\u63a5\u5bf9\u56fe\u50cf\u5168\u5c40\u8fdb\u884c\u68c0\u6d4b(\u5206\u5757\u6548\u7387\u4f4e\uff0c\u8ba1\u7b97\u5197\u4f59)\u3002<\/p>\n

\uff082\uff09\u9488\u5bf9\u9065\u611f\u56fe\u50cf\u4e2d\u76ee\u6807\u5bc6\u96c6\u5206\u5e03\u548c\u5916\u89c2\u6a21\u7cca\u7b49\u5e38\u89c1\u7279\u70b9, \u76ee\u524d\u8fd8\u6ca1\u6709\u8f83\u597d\u7684\u9488\u5bf9\u6027\u5904\u7406\u65b9\u5f0f\u3002<\/p>\n

\uff083\uff09\u73b0\u6709\u76ee\u6807\u68c0\u6d4b\u65b9\u6cd5\u5927\u591a\u662f\u57fa\u4e8e\u76ee\u6807\u672c\u8eab\u89c6\u89c9\u7279\u5f81, \u7f3a\u5c11\u6839\u636e\u56fe\u50cf\u6574\u4f53\u548c\u4e0a\u4e0b\u6587\u8fdb\u884c\u7406\u89e3\u548c\u63a8\u7406\u7684\u8fc7\u7a0b\uff08\u7f3a\u4e4f\u9ad8\u5c42\u8bed\u4e49\u77e5\u8bc6\u5f15\u5bfc\uff09<\/p>\n

\uff084\uff09\u73b0\u6709\u6570\u636e\u96c6 \u7684\u8d28\u91cf\u8fd8\u6709\u5f85\u63d0\u5347, \u5927\u91cf\u5b9e\u9645\u5b58\u5728\u7684\u5c0f\u76ee\u6807\u3001\u6a21\u7cca\u76ee\u6807\u6ca1\u6709\u6807\u6ce8\u51fa\u6765, \u9650\u5236\u4e86\u73b0\u6709\u7b97\u6cd5\u6f5c\u529b\u3002<\/p>\n

\uff085\uff09\u65cb\u8f6c\u6846\u66f4\u7d27\u51d1\uff0c\u4f46\u662f\u76f8\u5bf9\u4e8e\u6c34\u5e73\u6846\u7cbe\u5ea6\u6bd4\u8f83\u4f4e\uff0c\u5b9a\u4f4d\u4e0d\u51c6\u786e\u3002<\/p>\n

\uff086\uff09\u9488\u5bf9\u9065\u611f\u56fe\u50cf\u6570\u636e\u89c4\u6a21\u8f83\u5927\u7684\u95ee\u9898, \u73b0\u6709\u65b9 \u6cd5\u7684\u5904\u7406\u901f\u5ea6\u8f83\u6162\u3002<\/p>\n

5.2 \u672a\u6765\u8d8b\u52bf<\/h3>\n

\uff081\uff09\u91c7\u7528\u57fa\u4e8e\u56fe\u50cf\u6574\u4f53\u8fdb\u884c\u611f\u5174\u8da3\u533a\u57df\u63d0\u53d6\u7684\u65b9\u5f0f\u6765\u66ff\u4ee3\u5206\u5757 \u5904\u7406\u65b9\u5f0f, \u53ef\u4ee5\u5feb\u901f\u6ee4\u9664\u5927\u90e8\u5206\u80cc\u666f\u533a\u57df, \u4ece\u800c\u907f\u514d\u8ba1\u7b97\u5197\u4f59, \u63d0\u9ad8\u7b97\u6cd5\u6548\u7387\u3002<\/p>\n

\uff082\uff09\u52a0\u5f3a\u5c0f\u76ee\u6807\u3001 \u5bc6\u96c6\u76ee\u6807\u3001\u6a21\u7cca\u76ee\u6807\u7b49\u8f83\u96be\u68c0\u6d4b\u76ee\u6807\u7684\u7279\u5f81\u8868\u793a\uff0c\u964d\u4f4e\u80cc\u666f\u5e72\u6270\u3002<\/p>\n

\uff083\uff09\u5f15\u5165\u77e5\u8bc6\u548c\u63a8\u7406\u6a21\u5757 (\u4f8b \u5982\u77e5\u8bc6\u56fe\u8c31\u3001\u56fe\u5377\u79ef\u7b49) \u6765\u8f85\u52a9\u8fdb\u884c\u76ee\u6807\u68c0\u6d4b\u3002<\/p>\n

\uff084\uff09\u91c7\u7528\u5f31\u76d1\u7763\u5b66\u4e60\u3001\u534a\u76d1\u7763\u5b66\u4e60\u3001\u8fc1\u79fb\u5b66\u4e60\u7b49\u6765\u89e3\u51b3\u6807\u6ce8\u7684\u8d28\u91cf\u7684\u95ee\u9898<\/p>\n

\uff085\uff09\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u4e2d\u65cb\u8f6c\u6846\u5177\u6709\u66f4\u52a0\u7cbe\u7ec6\u7684\u5b9a\u4f4d\u80fd\u529b, \u4ee3\u66ff\u6c34\u5e73\u6846\u8fdb\u884c\u68c0\u6d4b\u662f\u5927\u52bf\u6240\u8d8b\uff0c\u8fdb\u4e00\u6b65\u63d0\u9ad8\u5b9a\u4f4d\u7684\u80fd\u529b\u3002<\/p>\n

\uff086\uff09\u8f7b\u91cf\u5316\u7f51\u7edc\u7684\u7ed3\u6784\u66f4\u5177\u6709\u91cd\u8981\u610f\u4e49\u4e0e\u4ef7\u503c\u3002\u7f51\u7edc\u67b6\u6784\u641c\u7d22\uff08NAS\uff09\u4e5f\u5177\u6709\u6f5c\u529b\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"

\u5149\u5b66\u9065\u611f\u56fe\u50cf\u76ee\u6807\u68c0\u6d4b\u7b97\u6cd5\u7efc\u8ff0\uff08\u4e09\uff09 \u539f\u6587pdf\u4e0b\u8f7d\uff1a\u4e0b\u8f7d\u94fe\u63a5 4. \u6570\u636e\u96c6\u548c\u7b97\u6cd5\u6027\u80fd\u6bd4\u8f83 4.1 \u9065\u611f\u56fe\u50cf\u76ee\u6807… » \u9605\u8bfb\u5168\u6587<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-88","post","type-post","status-publish","format-standard","hentry","category-dl"],"_links":{"self":[{"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/posts\/88","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/comments?post=88"}],"version-history":[{"count":4,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/posts\/88\/revisions"}],"predecessor-version":[{"id":100,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/posts\/88\/revisions\/100"}],"wp:attachment":[{"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/media?parent=88"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/categories?post=88"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/apifj.com\/index.php\/wp-json\/wp\/v2\/tags?post=88"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}