{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T06:58:41Z","timestamp":1777705121037,"version":"3.51.4"},"reference-count":23,"publisher":"SAGE Publications","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IFS"],"published-print":{"date-parts":[[2021,9,15]]},"abstract":"<jats:p>Deep convolutional neural network (CNN) is difficult to deploy to mobile and portable devices due to its large number of parameters and floating-point operations (FLOPs). To tackle this problem, we propose a novel channel pruning method. We use the modified squeeze-and-excitation blocks (MSEB) to measure the importance of the channels in the convolutional layers. The unimportant channels, including convolutional kernels related to them, are pruned directly, which greatly reduces the storage cost and the number of calculations. For ResNet with basic blocks, we propose an approach to consistently prune all residual blocks in the same stage to ensure that the compact network structure is dimensionally correct. After pruning we retrain the compact network from scratch to restore the accuracy. Finally, we verify our method on CIFAR-10, CIFAR-100 and ILSVRC-2012. The results indicate that the performance of the compact network is better than the original network when the pruning rate is small. Even when the pruning amplitude is large, the accuracy can also be maintained or decreased slightly. On the CIFAR-100, when reducing the parameters and FLOPs up to 82% and 62% respectively, the accuracy of VGG-19 even improve by 0.54% after retraining. The source code is available at https:\/\/github.com\/JingfeiChang\/UCP.<\/jats:p>","DOI":"10.3233\/jifs-202290","type":"journal-article","created":{"date-parts":[[2021,7,30]],"date-time":"2021-07-30T10:54:58Z","timestamp":1627642498000},"page":"2687-2699","source":"Crossref","is-referenced-by-count":1,"title":["Compressing convolutional neural networks via intermediate features"],"prefix":"10.1177","volume":"41","author":[{"given":"Chang","family":"Jingfei","sequence":"first","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lu","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"},{"name":"Anhui Mine IOT and Security Monitoring Technology Key Laboratory, Hefei, China"},{"name":"Engineering Research Center of Safety Critical Industrial Measurement and Control Technology, Ministry of Education, Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xue","family":"Ping","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wei","family":"Xing","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wei","family":"Zhen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"},{"name":"Anhui Mine IOT and Security Monitoring Technology Key Laboratory, Hefei, China"},{"name":"Engineering Research Center of Safety Critical Industrial Measurement and Control Technology, Ministry of Education, Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"issue":"7","key":"10.3233\/JIFS-202290_ref1","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1162\/neco.2006.18.7.1527","article-title":"A fast learning algorithmfor deep belief nets","volume":"18","author":"Hinton","year":"2006","journal-title":"Neural Computation"},{"issue":"5786","key":"10.3233\/JIFS-202290_ref2","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality ofdata with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"issue":"7553","key":"10.3233\/JIFS-202290_ref3","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"10.3233\/JIFS-202290_ref4","doi-asserted-by":"crossref","unstructured":"LeCun Y. , Kavukcuoglu K. and Farabet C. , \u201cConvolutional networks and applications in vision,\u201d in 2010 IEEE International Symposium on Circuits and Systems, 2010, pp. 253\u2013256.","DOI":"10.1109\/ISCAS.2010.5537907"},{"key":"10.3233\/JIFS-202290_ref5","first-page":"2278","article-title":"Gradientbasedlearning applied to document recognition","volume":"86","author":"Lecun","year":"1998","journal-title":"Proceedings of theIEEE"},{"key":"10.3233\/JIFS-202290_ref6","first-page":"84","article-title":"Imagenet classificationwith deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Communications of theACM"},{"key":"10.3233\/JIFS-202290_ref7","first-page":"818","article-title":"\u201cVisualizing and understandingconvolutional networks,\u201d in","volume":"8689","author":"Zeiler","year":"2014","journal-title":"ECCV 2014"},{"issue":"1","key":"10.3233\/JIFS-202290_ref18","doi-asserted-by":"crossref","first-page":"409","DOI":"10.3233\/JIFS-179096","article-title":"Accelerating convolutionalneural network-based malware traffic detection through ant-colonyclustering","volume":"37","author":"Huang","year":"2019","journal-title":"Journal of Intelligent & Fuzzy Systems"},{"issue":"10","key":"10.3233\/JIFS-202290_ref19","doi-asserted-by":"crossref","first-page":"2525","DOI":"10.1109\/TPAMI.2018.2858232","article-title":"\u201cThinet: Pruning cnn filters for a thinner net,\u201d","volume":"41","author":"Luo","year":"2019","journal-title":"IEEETranscations on Pattern Analysis and Machine Intelligence"},{"key":"10.3233\/JIFS-202290_ref20","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1016\/j.neucom.2019.08.063","article-title":"Retraining-freemethods for fast on-the-fly pruning of convolutional neuralnetworks","volume":"370","author":"Ashouri","year":"2019","journal-title":"Neurocomputing"},{"issue":"3","key":"10.3233\/JIFS-202290_ref21","doi-asserted-by":"crossref","first-page":"4003","DOI":"10.3233\/JIFS-190184","article-title":"Deep pruned nets for efficient image-based plants diseaseclassification","volume":"37","author":"Too","year":"2019","journal-title":"Journal of Intelligent & Fuzzy Systems"},{"key":"10.3233\/JIFS-202290_ref25","first-page":"45","article-title":"Structured deep fisher pruning forefficient facial trait classification","volume":"77","author":"Tian","year":"2018","journal-title":"Image and VisionComputing"},{"key":"10.3233\/JIFS-202290_ref28","doi-asserted-by":"crossref","unstructured":"Singh P. , Kadi V.S.R. and Namboodiri V.P. , Falf convnets: Fatuousauxiliary loss based filter-pruning for efficient deep cnns, Imageand Vision Computing, 93 (2020), 103857.","DOI":"10.1016\/j.imavis.2019.103857"},{"issue":"2","key":"10.3233\/JIFS-202290_ref29","doi-asserted-by":"crossref","first-page":"2243","DOI":"10.3233\/JIFS-182544","article-title":"\u201cNeural network pruning based on input importance,\u201d","volume":"37","author":"Hewahi","year":"2019","journal-title":"Journal of Intelligent & Fuzzy Systems"},{"issue":"5","key":"10.3233\/JIFS-202290_ref31","doi-asserted-by":"crossref","first-page":"7403","DOI":"10.3233\/JIFS-200771","article-title":"Multi-granularity pruning for deep residualnetworks","volume":"39","author":"Huang","year":"2020","journal-title":"Journal of Intelligent & Fuzzy Systems"},{"issue":"1","key":"10.3233\/JIFS-202290_ref32","doi-asserted-by":"crossref","first-page":"221","DOI":"10.3233\/JIFS-191014","article-title":"A CNN channel pruning low-bit framework using weight quantization withsparse group lasso regularization","volume":"39","author":"Long","year":"2020","journal-title":"Journal of Intelligent &Fuzzy Systems"},{"key":"10.3233\/JIFS-202290_ref33","first-page":"107461","article-title":"Autopruner: An end-to-end trainable filterpruning method for efficient deep model inference","volume":"107","author":"Luo","year":"2020","journal-title":"PatternRecognition"},{"key":"10.3233\/JIFS-202290_ref35","unstructured":"Liu Z. , Sun M. , Zhou T. , Huang G. and Darrell T. , \u201cRethinking the value of network pruning,\u201d in International Conference on Learning Representations, 2019."},{"issue":"5","key":"10.3233\/JIFS-202290_ref36","doi-asserted-by":"crossref","first-page":"1135","DOI":"10.1587\/transinf.2019EDP7177","article-title":"Neural behavior-basedapproach for neural network pruning","volume":"E103D","author":"Kamma","year":"2020","journal-title":"IEICE Transactions onInformation and Systems"},{"issue":"5","key":"10.3233\/JIFS-202290_ref37","doi-asserted-by":"crossref","first-page":"1190","DOI":"10.1587\/transinf.2019EDL8200","article-title":"Loss-driven channelpruning of convolutional neural networks","volume":"E103D","author":"Long","year":"2020","journal-title":"IEICE Transactions onInformation and Systems"},{"key":"10.3233\/JIFS-202290_ref39","doi-asserted-by":"crossref","unstructured":"Lin M. , Ji R. , Zhang Y. , Zhang B. , Wu Y. and Tian Y. , \u201cChannel pruning via automatic structure search,\u201d in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2020, pp. 673\u2013679.","DOI":"10.24963\/ijcai.2020\/94"},{"key":"10.3233\/JIFS-202290_ref40","doi-asserted-by":"crossref","unstructured":"Yu R. , Li A. , Chen C. , Lai J. , Morariu V.I. , Han X. , Gao M. , Linand C. and Davis L.S. , \u201cNisp: Pruning networks using neuron importance score propagation,\u201d in 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 9194\u20139203.","DOI":"10.1109\/CVPR.2018.00958"},{"key":"10.3233\/JIFS-202290_ref41","doi-asserted-by":"crossref","unstructured":"Molchanov P. , Mallya A. , Tyree S. , Frosio I. and Kautz J. , \u201cImportance estimation for neural network pruning,\u201d in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019.","DOI":"10.1109\/CVPR.2019.01152"}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/JIFS-202290","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:42:57Z","timestamp":1777455777000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/JIFS-202290"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,15]]},"references-count":23,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.3233\/jifs-202290","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9,15]]}}}