{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T17:49:46Z","timestamp":1773942586576,"version":"3.50.1"},"reference-count":27,"publisher":"Wiley","license":[{"start":{"date-parts":[[2021,5,31]],"date-time":"2021-05-31T00:00:00Z","timestamp":1622419200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61673395"],"award-info":[{"award-number":["61673395"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Security and Communication Networks"],"published-print":{"date-parts":[[2021,5,31]]},"abstract":"<jats:p>Convolutional neural network (CNN) has been applied widely in various fields. However, it is always hindered by the unexplainable characteristics. Users cannot know why a CNN-based model produces certain recognition results, which is a vulnerability of CNN from the security perspective. To alleviate this problem, in this study, the three existing feature visualization methods of CNN are analyzed in detail firstly, and a unified visualization framework for interpreting the recognition results of CNN is presented. Here, class activation weight (CAW) is considered as the most important factor in the framework. Then, the different types of CAWs are further analyzed, and it is concluded that a linear correlation exists between them. Finally, on this basis, a spatial-channel attention-based class activation mapping (SCA-CAM) method is proposed. This method uses different types of CAWs as attention weights and combines spatial and channel attentions to generate class activation maps, which is capable of using richer features for interpreting the results of CNN. Experiments on four different networks are conducted. The results verify the linear correlation between different CAWs. In addition, compared with the existing methods, the proposed method SCA-CAM can effectively improve the visualization effect of the class activation map with higher flexibility on network structure.<\/jats:p>","DOI":"10.1155\/2021\/6682293","type":"journal-article","created":{"date-parts":[[2021,6,1]],"date-time":"2021-06-01T04:01:23Z","timestamp":1622520083000},"page":"1-13","source":"Crossref","is-referenced-by-count":5,"title":["Spatial-Channel Attention-Based Class Activation Mapping for Interpreting CNN-Based Image Classification Models"],"prefix":"10.1155","volume":"2021","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4619-4325","authenticated-orcid":true,"given":"Nianwen","family":"Si","sequence":"first","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6842-9722","authenticated-orcid":true,"given":"Wenlin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8876-5117","authenticated-orcid":true,"given":"Dan","family":"Qu","sequence":"additional","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3225-4649","authenticated-orcid":true,"given":"Xiangyang","family":"Luo","sequence":"additional","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8156-3347","authenticated-orcid":true,"given":"Heyu","family":"Chang","sequence":"additional","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0128-9657","authenticated-orcid":true,"given":"Tong","family":"Niu","sequence":"additional","affiliation":[{"name":"Information Engineering University, Zhengzhou 450001, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","reference":[{"key":"1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461972"},{"key":"4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N19-1423"},{"key":"5","first-page":"818","article-title":"Visualizing and understanding convolutional networks, computer vision\u2014ECCV 2014","author":"M. D. Zeiler"},{"key":"6","first-page":"97","article-title":"Why should I trust you?: explaining the predictions of any classifier","author":"M. T. Ribeiro"},{"key":"7","first-page":"8827","article-title":"Interpretable convolutional neural networks","author":"Q. Zhang"},{"key":"8","first-page":"2921","article-title":"Learning deep features for discriminative localization","author":"B. Zhou"},{"key":"9","first-page":"618","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","author":"R. R. Selvaraju"},{"key":"10","article-title":"Object detectors emerge in deep scene CNNS","author":"B. Zhou"},{"key":"11","article-title":"Deep inside convolutional networks: visualising image classification models and saliency maps","author":"K. Simonyan"},{"key":"12","first-page":"2640","article-title":"Rating-boosted latent topics: understanding users and items with ratings and reviews","author":"Y. Tan"},{"key":"13","first-page":"3549","article-title":"MDNet: a semantically and visually interpretable medical image diagnosis network","author":"Z. Zhang"},{"key":"14","first-page":"2961","article-title":"Interpretable learning for self-driving cars by visualizing causal attention","author":"J. Kim"},{"key":"15","first-page":"1","article-title":"Interpretable end-to-end urban autonomous driving with latent deep reinforcement learning","author":"J. Chen"},{"key":"16","first-page":"1106","article-title":"Imagenet classification with deep convolutional neural networks","author":"A. Krizhevsky"},{"key":"17","doi-asserted-by":"publisher","DOI":"10.3934\/mfc.2018008"},{"key":"18","article-title":"Very deep convolutional networks for large-scale image recognition","author":"K. Simonyan"},{"key":"19","first-page":"2261","article-title":"Densely connected convolutional networks","author":"G. Huang"},{"key":"20","first-page":"2818","article-title":"Rethinking the inception architecture for computer vision","author":"C. Szegedy"},{"key":"21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01234-2_1"},{"key":"22","first-page":"6298","article-title":"Spatial and channel-wise attention in convolutional networks for image captioning","author":"L. Chen"},{"key":"23","article-title":"Automatic differentiation in pytorch","author":"A. Paszke"},{"key":"24","first-page":"1485","article-title":"Torchvision the machine-vision package of torch","author":"S. Marcel"},{"key":"25","article-title":"Squeezenet: alexnet-level accuracy with 50x fewer parameters and <0.5\u2009MB model size","author":"F. N. Iandola"},{"key":"26","first-page":"248","article-title":"ImageNet: a large-scale hierarchical image database","author":"J. Deng"},{"key":"27","volume-title":"Tiny Imagenet Visual Recognition Challenge","author":"F. Li","year":"2015"}],"container-title":["Security and Communication Networks"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2021\/6682293.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2021\/6682293.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/scn\/2021\/6682293.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,6,1]],"date-time":"2021-06-01T04:01:27Z","timestamp":1622520087000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.hindawi.com\/journals\/scn\/2021\/6682293\/"}},"subtitle":[],"editor":[{"given":"Yuewei","family":"Dai","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2021,5,31]]},"references-count":27,"alternative-id":["6682293","6682293"],"URL":"https:\/\/doi.org\/10.1155\/2021\/6682293","relation":{},"ISSN":["1939-0122","1939-0114"],"issn-type":[{"value":"1939-0122","type":"electronic"},{"value":"1939-0114","type":"print"}],"subject":[],"published":{"date-parts":[[2021,5,31]]}}}