{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T21:24:35Z","timestamp":1784841875731,"version":"3.55.0"},"reference-count":26,"publisher":"Institution of Engineering and Technology (IET)","issue":"5","license":[{"start":{"date-parts":[[2019,3,20]],"date-time":"2019-03-20T00:00:00Z","timestamp":1553040000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/2.zoppoz.workers.dev:443\/http\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["ietresearch.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["IET Image Processing"],"published-print":{"date-parts":[[2019,4]]},"abstract":"<jats:p>Deep learning has achieved state\u2010of\u2010the\u2010art performance in accuracy of many computer vision tasks. However, convolutional neural network is difficult to deploy on resource constrained devices due to their limited computation power and memory space. Thus, it is necessary to prune the redundant weights and filters rationally and effectively. Considering that the pruned model still exists, redundancy after weight pruning or filter pruning alone, a method of combining weight pruning and filter pruning is proposed. First, filter pruning is performed, which is to remove filters with least importance and using fine\u2010tuning to recover the model's accuracy. Then, all connection weights below a threshold are set to zero. Finally, the pruned model obtained by the first two steps is fine\u2010tuned to recover its predictive accuracy. Experiments on MNIST and CIFAR\u201010 datasets demonstrate that the proposed approach is effective and feasible. Compared with only weight pruning or filter pruning, the mixed pruning can achieve higher compression ratio of the model parameters. For LeNet\u20105, the proposed approach can achieve a compression rate of 13.01\u00d7, with 1% drop in accuracy. For VGG\u201016, it can achieve a compression rate of 19.20\u00d7, incurring 1.56% accuracy loss.<\/jats:p>","DOI":"10.1049\/iet-ipr.2018.6191","type":"journal-article","created":{"date-parts":[[2019,1,18]],"date-time":"2019-01-18T21:37:13Z","timestamp":1547847433000},"page":"779-784","update-policy":"https:\/\/2.zoppoz.workers.dev:443\/https\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":20,"title":["Thinning of convolutional neural network with mixed pruning"],"prefix":"10.1049","volume":"13","author":[{"given":"Wenzhu","family":"Yang","sequence":"first","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lilei","family":"Jin","sequence":"additional","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sile","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenchao","family":"Cu","sequence":"additional","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiangyang","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liping","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Cyber Security and Computer, Hebei University Baoding People's Republic of China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"265","published-online":{"date-parts":[[2019,3,20]]},"reference":[{"key":"e_1_2_7_2_1","unstructured":"2012 Proc. Int. Conf. Neural Information Processing Systems (NIPS) Nevada USA A. Krizhevsky I. Sutskever G.E. Hinton Imagenet classification with deep convolutional neural networks 1097 1105"},{"key":"e_1_2_7_3_1","unstructured":"2015 Proc. Int. Conf. on Learning Representation (ICLR) San Diego USA K. Simonyan A. Zisserman Very deep convolutional networks for large\u2010scale image recognition"},{"key":"e_1_2_7_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.169"},{"key":"e_1_2_7_5_1","doi-asserted-by":"publisher","DOI":"10.1049\/iet-ipr.2017.0070"},{"key":"e_1_2_7_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_7_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"e_1_2_7_8_1","unstructured":"2016 Proc. Int. Conf. Computer Vision and Pattern Recognition (CVPR) Las Vegas USA G. Huang Z. Liu V.D.M. Laurens Densely connected convolutional networks 2261 2269"},{"key":"e_1_2_7_9_1","unstructured":"1990 Proc. Int. Conf. Neural Information Processing Systems (NIPS) Denver CO USA Y. LeCun J.S. Denker S.A. Solla Optimal brain damage 598 605"},{"key":"e_1_2_7_10_1","unstructured":"2015 Proc. Int. Conf. Neural Information Processing Systems (NIPS) Montreal Canada S. Han J. Pool J. Tran Learning both weights and connections for efficient neural network 1135 1143"},{"key":"e_1_2_7_11_1","unstructured":"2013 Proc. Int. Conf. Neural Information Processing Systems (NIPS) Lake Tahoe USA M. Denil B. Shakibi L. Dinh Predicting parameters in deep learning 2148 2156"},{"key":"e_1_2_7_12_1","unstructured":"IandolaF.N. HanS. andMoskewiczM.W.et al: \u2018Squeezenet: AlexNet\u2010level accuracy with 50x fewer parameters and <0.5\u00a0MB model size\u2019 2016 arXiv:1602.07360"},{"key":"e_1_2_7_13_1","unstructured":"HowardA.G. ZhuM. andChenB.et al: \u2018Mobilenets: efficient convolutional neural networks for mobile vision applications\u2019 2017 arXiv:1704.04861"},{"key":"e_1_2_7_14_1","doi-asserted-by":"crossref","unstructured":"ZhangX. ZhouX. andLinM.et al: \u2018Shufflenet: an extremely efficient convolutional neural network for mobile devices\u2019 2017 arXiv:1707.01083","DOI":"10.1109\/CVPR.2018.00716"},{"key":"e_1_2_7_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2015.07.093"},{"issue":"7","key":"e_1_2_7_16_1","first-page":"38","article-title":"Distilling the knowledge in a neural network","volume":"14","author":"Hinton G.","year":"2015","journal-title":"Comput. Sci."},{"key":"e_1_2_7_17_1","unstructured":"SauB.B. andBalasubramanianV.N.: \u2018Deep model compression: distilling knowledge from noisy teachers\u2019 2016 arXiv:1610.09650"},{"key":"e_1_2_7_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.521"},{"key":"e_1_2_7_19_1","unstructured":"2016 Proc. Int. Conf. on Learning Representation (ICLR) San Juan Argentina S. Han H. Mao W.J. Dally Deep compression: compressing deep neural networks with pruning trained quantization and huffman coding 3 7"},{"key":"e_1_2_7_20_1","unstructured":"2017 Proc. Int. Conf. on Learning Representation (ICLR) Toulon France P. Molchanov S. Tyree T. Karras Pruning convolutional neural networks for resource efficient transfer learning 1 17"},{"key":"e_1_2_7_21_1","unstructured":"2017 Proc. Int. Conf. on Learning Representation (ICLR) Toulon France H. Li A. Kadav I. Durdanovic Pruning filters for efficient ConvNets"},{"key":"e_1_2_7_22_1","unstructured":"HuH. PengR. andTaiY.W.et al: \u2018Network trimming: a data\u2010driven neuron pruning approach towards efficient deep architectures\u2019 2016 arXiv:1607.03250"},{"key":"e_1_2_7_23_1","unstructured":"LuoJ.H. andWuJ.: \u2018An entropy\u2010based pruning method for CNN compression\u2019 2017 arXiv:1706.05791"},{"key":"e_1_2_7_24_1","unstructured":"2017 Proc. Int. Conf. on Computer Vision (ICCV) Venice Italy J.H. Luo J.X. Wu W.Y. Lin Thinet: A filter level pruning method for deep neural network compression 5068 5076"},{"key":"e_1_2_7_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_7_26_1","unstructured":"2009 University of Toronto Canada A. Krizhevsky Learning multiple layers of features from tiny images 1 60"},{"key":"e_1_2_7_27_1","unstructured":"2017 Proc. Int. Conf. Neural Information Processing Systems (NIPS) Long Beach USA A. Paszke S. Gross S. Chintala Automatic differentiation in PyTorch 1 4"}],"container-title":["IET Image Processing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/iet-ipr.2018.6191","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1049\/iet-ipr.2018.6191","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/ietresearch.onlinelibrary.wiley.com\/doi\/pdf\/10.1049\/iet-ipr.2018.6191","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T05:56:33Z","timestamp":1761630993000},"score":1,"resource":{"primary":{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/ietresearch.onlinelibrary.wiley.com\/doi\/10.1049\/iet-ipr.2018.6191"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,3,20]]},"references-count":26,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2019,4]]}},"alternative-id":["10.1049\/iet-ipr.2018.6191"],"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/doi.org\/10.1049\/iet-ipr.2018.6191","archive":["Portico"],"relation":{},"ISSN":["1751-9659","1751-9667"],"issn-type":[{"value":"1751-9659","type":"print"},{"value":"1751-9667","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,3,20]]},"assertion":[{"value":"2018-09-12","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-01-06","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-03-20","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}