|本期目录/Table of Contents|

[1]吕秀丽,黄兆昊,白永强.基于改进LBP和DBN的人脸识别算法研究[J].工业仪表与自动化装置,2021,(05):80-82.[doi:10.19950/j.cnki.cn61-1121/th.2021.05.017]
 L Xiuli,HUANG Zhaohao,BAI Yongqiang.Study on face recognition algorithm based on improved LBP and DBN[J].Industrial Instrumentation & Automation,2021,(05):80-82.[doi:10.19950/j.cnki.cn61-1121/th.2021.05.017]
点击复制

基于改进LBP和DBN的人脸识别算法研究

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:
期数:
2021年05期
页码:
80-82
栏目:
出版日期:
2021-10-15

文章信息/Info

Title:
Study on face recognition algorithm based on improved LBP and DBN
作者:
吕秀丽黄兆昊白永强
东北石油大学 物理与电子工程学院,黑龙江 大庆 163318
Author(s):
L XiuliHUANG ZhaohaoBAI Yongqiang
(Northeast Petroleum University,School of Physics and Electronic Engineering,Heilongjiang Daqing 163318,China)
关键词:
局部二值模式纹理特征特征融合深度信念网络
Keywords:
local binary patterntexture featurefeature fusiondepth belief network
分类号:
TP391.41
DOI:
10.19950/j.cnki.cn61-1121/th.2021.05.017
文献标志码:
A
摘要:
针对传统LBP算法的人脸识别易受光照、背景、遮挡等因素的影响,使用改进局部二值模式(LBP)和深度信念网络(DBN)相结合的方法,用多尺度块局部二值模式(MB-LBP)算法获取人脸图像的纹理特征,在此人脸纹理特征的基础上使用中心对称局部二值模式(CS-LBP)算法获取图像的纹理特征,然后将两次获得的纹理特征图像的直方图进行融合,并将其输入到DBN中进行训练,优化网络参数。
Abstract:
The traditional LBP algorithm for face recognition is easily affected by illumination, background, occlusion and other factors. Using the improved local binary pattern (LBP) and depth belief network (DBN) method, the texture features of face image are obtained by using multi-scale block local binary pattern (MB-LBP) algorithm. On the basis of the face texture features, the centrosymmetric local binary pattern (CS-LBP) is used the algorithm obtains the texture features of the image, and then fuses the histograms of the two texture feature images, and inputs them into the DBN for training to optimize the network parameters.

参考文献/References:

[1]ANBARJAFARI G. Face recognition using color local binary pattern from mutually independent color channels [J].Eurasip Journal n Image & Video processing, 2013(1):4049-4068.

[2]TORTORICT,WERGHI N, BERRETTI S. Boosting 3D LBP-based face recognition by fusing shape and texture descriptors on the mesh[C]//IEEE International Conference on Image Processing, IEEE,2015:964-979.
[3]杨恢先,贺迪龙,刘凡,等.基于双向梯度中心对称局部二值模式的单样本人脸识别[J].计算机辅助设计与图形学报,2017,29(1):130-136.
[4]胡敏,余子玺,王晓华,等.G-LBP和方差投影交叉熵的人脸识[J].图学学报,2017,38(1):82-89.
[5]CHEN W S, DAI X, PAN B, et al. A novel discriminant criterion based on feature fusion strategy for face recognition[ J ]. Neurocomp-uting,2015,159(1):67-77.
[6]卢建云,何中市,余磊.基于多级CS-LBP特征融合的人脸识别方法[J].计算机工程与科学,2010,32(6):48-51.
[7]刘斌,米强,徐岩.LBP和MB-LBP加权融合的人脸识别[J].计算机工程与设计,2018,39(02):551-556.
[8]张敦凤,高宁化,王姮,等.基于分块LBP融合特征和SVM的人脸识别算法[J].传感器与微系统,2019,38(5):154-156+160.
[9]REN J , JIANG X , YUAN J . Quantized fuzzy LBP for face recognition[J]. IEEE, 2015:1503-1507.
[10]苏育挺,张天娇,张静,等.基于局部二值模式的中值滤波检测算法[J].计算机应用研究,2016,33(01):258-261.
[11]刘方园,王水花,张煜东.深度置信网络模型及应用研究综述[J].计算机工程与应用,2018,54(01):11-18+47.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2021-04-07

基金项目:
黑龙江省自然科学基金资助(LH2019D006)

作者简介:
吕秀丽(1971),女,教授,博士,主要研究领域为数字图像处理、生物特征识别技术、数字水印与信息隐藏。
更新日期/Last Update: 1900-01-01