{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T07:11:42Z","timestamp":1761635502794,"version":"build-2065373602"},"reference-count":36,"publisher":"Institution of Engineering and Technology (IET)","issue":"2","license":[{"start":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T00:00:00Z","timestamp":1635724800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["ietresearch.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["IET Computer Vision"],"published-print":{"date-parts":[[2022,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The loss function is a criterion to evaluate the learning quality of a deep convolutional neural network, which represents the gap between prediction and ground truth. However, as the most commonly used loss function in image classification tasks, Cross\u2010Entropy loss does not encourage the model to distinguish the similarity between features. In this work, the authors investigate inter\u2010class separability of similar features learnt by convolutional networks and propose a loss function called Error Refactor Loss (ER\u2010Loss). ER\u2010Loss is based on the error caused by convolutional networks; it can improve the inter\u2010class separability and is simple to implement and can easily replace the Cross\u2010Entropy loss. Compared with softmax loss, ER\u2010Loss adds a dynamic penalty item which can help ER\u2010Loss monitor the actual situation of model training and adjust the value of the penalty item according to model training. The ER\u2010Loss on CIFAR100 and part of ImageNet ILSVRC 2012 is evaluated and the experimental result showed that the ER\u2010Loss can improve the accuracy of the model.<\/jats:p>","DOI":"10.1049\/cvi2.12079","type":"journal-article","created":{"date-parts":[[2021,11,1]],"date-time":"2021-11-01T08:58:23Z","timestamp":1635757103000},"page":"192-203","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Error Refactor loss based on error analysis in image classification"],"prefix":"10.1049","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2255-275X","authenticated-orcid":false,"given":"Xiaoyu","family":"Yu","sequence":"first","affiliation":[{"name":"College of Electron and Information University of Electronic Science and Technology of China Zhongshan Institute  Zhongshan China"}]},{"given":"Yinglu","family":"Chen","sequence":"additional","affiliation":[{"name":"College of Electron and Information University of Electronic Science and Technology of China Zhongshan Institute  Zhongshan China"},{"name":"South China Academy of Advanced Optoelectronics South China Normal University  Guangzhou China"}]},{"given":"Guofu","family":"Zhou","sequence":"additional","affiliation":[{"name":"South China Academy of Advanced Optoelectronics South China Normal University  Guangzhou China"}]},{"given":"Yan","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Electron and Information University of Electronic Science and Technology of China Zhongshan Institute  Zhongshan China"}]},{"given":"Fuchao","family":"Li","sequence":"additional","affiliation":[{"name":"College of Electron and Information University of Electronic Science and Technology of China Zhongshan Institute  Zhongshan China"},{"name":"South China Academy of Advanced Optoelectronics South China Normal University  Guangzhou China"}]},{"given":"Zhifei","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Linguistics University of Washington  Seattle WA USA"}]}],"member":"265","published-online":{"date-parts":[[2021,11]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.5555\/3042817.3043008"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"volume-title":"Building and Leveraging Category Hierarchies for Large\u2010scale Image Classification","author":"Hao Z.","key":"e_1_2_10_4_1"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/WACV.2019.00073"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tplants.2014.08.004"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1002\/j.1538\u20107305.1974.tb02756.x"},{"key":"e_1_2_10_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jfineco.2003.02.001"},{"key":"e_1_2_10_9_1","doi-asserted-by":"publisher","DOI":"10.1049\/ccs.2019.0010"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107405"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/LGRS.2019.2937872"},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2019.2909737"},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.rse.2006.06.010"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2006.100"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298682"},{"volume-title":"In Defense of the Triplet Loss for Person Re\u2010Identifications","year":"2017","author":"Hermans A.","key":"e_1_2_10_16_1"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.145"},{"volume-title":"Deep Learning Using Linear Support Vector Machines","year":"2013","author":"Tang Y.","key":"e_1_2_10_18_1"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1007\/978\u20103\u2010319\u201046478\u2010731"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1088\/1742-6596\/1229\/1\/012006"},{"key":"e_1_2_10_21_1","doi-asserted-by":"publisher","DOI":"10.5555\/2627435.2670313"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1007\/978\u20103\u2010319\u201070096\u20100_43"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF00058611"},{"key":"e_1_2_10_24_1","first-page":"1251","volume-title":"Proceedings of the 24th International Conference on Neural Information Processing Systems (NIPS)","author":"Zhao B.","year":"2011"},{"volume-title":"Making Better Mistakes: Leveraging Class Hierarchies With Deep Networks","year":"2019","author":"Bertinetto L.","key":"e_1_2_10_25_1"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.4140\/TCP.n.2015.249"},{"key":"e_1_2_10_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.2205597"},{"key":"e_1_2_10_28_1","doi-asserted-by":"publisher","DOI":"10.9774\/GLEAF.978\u20101\u2010909493\u201038\u20104_2"},{"key":"e_1_2_10_29_1","doi-asserted-by":"publisher","DOI":"10.5120\/13897\u20101851"},{"key":"e_1_2_10_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.474"},{"key":"e_1_2_10_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2913372"},{"key":"e_1_2_10_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00060"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"e_1_2_10_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"e_1_2_10_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_10_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"}],"container-title":["IET Computer Vision"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/cvi2.12079","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1049\/cvi2.12079","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/cvi2.12079","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T07:05:49Z","timestamp":1761635149000},"score":1,"resource":{"primary":{"URL":"https:\/\/ietresearch.onlinelibrary.wiley.com\/doi\/10.1049\/cvi2.12079"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,11]]},"references-count":36,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2022,3]]}},"alternative-id":["10.1049\/cvi2.12079"],"URL":"https:\/\/doi.org\/10.1049\/cvi2.12079","archive":["Portico"],"relation":{},"ISSN":["1751-9632","1751-9640"],"issn-type":[{"type":"print","value":"1751-9632"},{"type":"electronic","value":"1751-9640"}],"subject":[],"published":{"date-parts":[[2021,11]]},"assertion":[{"value":"2020-12-30","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-10-05","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-11-01","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}