{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T14:17:05Z","timestamp":1740147425226,"version":"3.37.3"},"reference-count":22,"publisher":"Wiley","license":[{"start":{"date-parts":[[2021,4,7]],"date-time":"2021-04-07T00:00:00Z","timestamp":1617753600000},"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":["61741124","5781"],"award-info":[{"award-number":["61741124","5781"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Guizhou Science and Technology Plan","award":["61741124","5781"],"award-info":[{"award-number":["61741124","5781"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Advances in Multimedia"],"published-print":{"date-parts":[[2021,4,7]]},"abstract":"<jats:p>Aiming at the problem of a large number of parameters and high time complexity caused by the current deep convolutional neural network models, an improved face alignment algorithm of a cascaded convolutional neural network (CCNN) is proposed from the network structure, random perturbation factor (shake), and data scale. The algorithm steps are as follows: 3 groups of lightweight CNNs are designed; the first group takes facial images with face frame as input, trains 3 CNNs in parallel, and weighted outputs the facial images with 5 facial key points (anchor points). Then, the anchor points and 2 different windows with a shake mechanism are used to crop out 10 partial images of human faces. The networks in the second group train 10 CNNs in parallel and every 2 networks\u2019 weighted average and colocated a key point. Based on the second group of networks, the third group designed a smaller shake mechanism and windows, to achieve more fine-tuning. When training the network, the idea of parallel within groups and serial between groups is adopted. Experiments show that, on the LFPW face dataset, the improved CCNN in this paper is superior to any other algorithm of the same type in positioning speed, algorithm parameter amount, and test error.<\/jats:p>","DOI":"10.1155\/2021\/6654561","type":"journal-article","created":{"date-parts":[[2021,4,8]],"date-time":"2021-04-08T22:55:37Z","timestamp":1617922537000},"page":"1-9","source":"Crossref","is-referenced-by-count":2,"title":["Face Alignment Algorithm Based on an Improved Cascaded Convolutional Neural Network"],"prefix":"10.1155","volume":"2021","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7623-5816","authenticated-orcid":true,"given":"Xun","family":"Duan","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology, GuiZhou University, GuiYang 550025, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4517-2917","authenticated-orcid":true,"given":"Yuanshun","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, GuiZhou University, GuiYang 550025, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6782-5777","authenticated-orcid":true,"given":"Yun","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, GuiZhou University, GuiYang 550025, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","reference":[{"key":"1","first-page":"1001","article-title":"Face landmark detection network with multi-scale feature map fusion","volume":"1-6","author":"G. Q. Qi","year":"2019","journal-title":"Application Research of Computers"},{"issue":"06","key":"2","first-page":"434","article-title":"Simulation of face key point recognition and location method based on deep learning","volume":"37","author":"J. Xu","year":"2020","journal-title":"Computer Simulation"},{"volume-title":"The Research and Implement of Face Detection and Face Alignment in the Wild","year":"2019","author":"J. Li","key":"3"},{"volume-title":"Research on Facial Landmark Localization Based on Deep Learning","year":"2019","author":"X. F. Yang","key":"4"},{"volume-title":"Research on Face Detection and Face Alignment Based on Deep Learning","year":"2018","author":"C. X. Jing","key":"5"},{"key":"6","doi-asserted-by":"publisher","DOI":"10.1006\/cviu.1995.1004"},{"author":"T. F. Cootes","key":"7","article-title":"Active appearance models"},{"issue":"11","key":"8","first-page":"228","article-title":"New neural network method for solving nonsmooth pseudoconvex optimization problems","volume":"46","author":"X. Yu","year":"2019","journal-title":"Computer Science"},{"key":"9","article-title":"Going deeper in spiking neural networks: VGG and residual rrchitectures","volume":"13","author":"A. Sengupta","year":"2018","journal-title":"Frontiers in Neuroence"},{"article-title":"Inception-v4, inception-resNet and the impact of residual connections on learning","year":"2016","author":"C. Szegedy","key":"10"},{"article-title":"Deep residual learning for image recognition","year":"2015","author":"K. He","key":"11"},{"key":"12","doi-asserted-by":"publisher","DOI":"10.19734\/j.issn.1001-3695.2019.05.0168"},{"key":"13","first-page":"32","article-title":"Face key point location based on cascaded convolutional neural network","volume":"30","author":"R. Chen","year":"2017","journal-title":"Journal of SiChuan Technology University (natural Edition)"},{"key":"14","article-title":"A high- precision and lightweight facial landmark detection algorithm","volume":"1-12","author":"L. H. Xu","year":"2020","journal-title":"Laser & Optoelectronics Progress"},{"issue":"01","key":"15","first-page":"24","article-title":"Structure optimization of convolutional neural networks:a survey","volume":"46","author":"J. D. Lin","year":"2020","journal-title":"Journal of Automatica Sinica"},{"issue":"02","key":"16","first-page":"1","article-title":"A review of object detection algorithms based on convolutional neural networks","volume":"37","author":"F. Y. Hu","year":"2020","journal-title":"Journal of SuZhou University of Science and Technology (Natural Science Edition)"},{"volume-title":"Research on Occlusion and Pose Robust Facial Landmark Localization","year":"2019","author":"S. Wu","key":"17"},{"issue":"01","key":"18","first-page":"93","article-title":"Adaptive task sorting model based on improved linear regression method","volume":"48","author":"Y. H. Wu","year":"2020","journal-title":"Journal of Huazhong University of Science and Technology (Natural Science Edition)"},{"key":"19","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume":"9","author":"X. Glorot","year":"2010","journal-title":"Journal of Machine Learning Research"},{"key":"20","doi-asserted-by":"crossref","DOI":"10.1016\/j.patrec.2020.06.015","article-title":"Data augmentation method for improving the accuracy of human pose estimation with cropped images","volume":"136","author":"S. Park","year":"2020","journal-title":"Pattern Recognition Letters"},{"issue":"01","key":"21","first-page":"22","article-title":"Power system transient stability prediction based on data augmentation and deep residual network","volume":"53","author":"Y. Z. Zhou","year":"2020","journal-title":"Journal of China Electric Power"},{"issue":"4","key":"22","first-page":"3476","article-title":"Deep convolutional network cascade for facial point detection","volume":"9","author":"Y. Sun","year":"2013","journal-title":"Computer Vision and Pattern Recognition"}],"container-title":["Advances in Multimedia"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/am\/2021\/6654561.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/am\/2021\/6654561.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/am\/2021\/6654561.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,4,8]],"date-time":"2021-04-08T22:56:00Z","timestamp":1617922560000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.hindawi.com\/journals\/am\/2021\/6654561\/"}},"subtitle":[],"editor":[{"given":"Deepu","family":"Rajan","sequence":"additional","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2021,4,7]]},"references-count":22,"alternative-id":["6654561","6654561"],"URL":"https:\/\/doi.org\/10.1155\/2021\/6654561","relation":{},"ISSN":["1687-5699","1687-5680"],"issn-type":[{"type":"electronic","value":"1687-5699"},{"type":"print","value":"1687-5680"}],"subject":[],"published":{"date-parts":[[2021,4,7]]}}}