基于深度学习算法的人脸图像活体特征变换尺度提取  

Adaptive blur and deduplication algorithm for digital media image based on wavelet domain

作　　者：吕晓琪[1,2,3] 李浩 谷宇 LYU Xiao-qi;LI Hao;GU Yu(College of Information Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China;Inner Mongolia Key Laboratory of Pattern Recognition and Intelligent Imag Processing,Inner Mongolia University of Science and Technology,Baotou 014010,China;Institute of Information Engineering,Inner Mongolia University of Technology,Hohhot 010051,China)

机构地区：[1]内蒙古科技大学信息工程学院,内蒙古包头014010 [2]内蒙古科技大学内蒙古自治区模式识别与智能图象处理重点实验室,内蒙古包头014010 [3]内蒙古工业大学信息工程学院,呼和浩特010051

出　　处：《吉林大学学报（工学版）》2023年第11期3201-3206,共6页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(62001255,61771266,61841204);内蒙古自治区科技计划项目(2019GG138);中央引导地方科技发展项目(2021ZY0004);内蒙古自治区自然科学基金项目(2019MS06003,2015MS0604);内蒙古自治区高等学校科学研究项目(NJZY145);教育部“春晖计划”合作科研项目(教外司留[2019]1383号).

摘　　要：针对目前人脸图像的特征提取方法存在提取准确度低、效率低等问题,提出了基于深度学习算法的人脸图像活体特征变换尺度提取方法。先利用深度学习方法完成人脸图像的去噪处理;然后使用Gabor波滤器分解人脸信号,并将其输入到深层子空间模型完成特征变换尺度的提取;最后基于(粒子PSO群优化算法)完成人脸图像活体特征变换尺度提取。实验结果表明:提出人脸图像特征提取方法的准确度更高,识别速度更快,整体应用效果更好。The quality of face feature extraction results affects the accuracy of face recognition.At present,the feature extraction methods of face images still have the problems of low extraction accuracy and low efficiency.In order to solve the problems in the methods,a scale extraction method of face image living feature transformation based on deep learning algorithm is proposed.The deep learning method is used to denoise the face image.Based on this,Gabor wave filter is used to decompose the face signal and input it into the deep subspace model to extract the feature transform scale.Based on PSO(particle swarm optimization),the scale extraction of face image living feature transformation is completed.The experimental results show that the proposed face image feature extraction method has higher accuracy,faster recognition speed and better overall application effect.

关 键 词：人脸特征提取 深度学习 去噪处理 Gabor波滤器 深层子空间模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]