基于加权LBP的自动人脸识别系统实现

摘要自动人脸识别已成为人工智能和模式识别领域的研究热点，在信息安全、自助服务、电子商务等多个领域有着众多应用。局部二值模式(Local Binary Pattern,LBP)是一种描述图像局部纹理特征的算子，具有运算速度快、灰度不变性和旋转不变性等优点。本课题实现了一个基于加权LBP的自动人脸识别系统，使用MFC做界面，借助于工具OpenCV实现人脸识别过程。首先在图像中检测人脸，将人脸区域划分成多个局部区域，对每个局部区域提取LBP纹理特征，并转化为统计直方图，通过直方图之间的卡方检验可以判断直方图之间的相似度，利用不同局部区域的权重，进而找到待识别图像的最佳匹配对象，完成人脸识别任务。82733

毕业论文关键词人脸识别局部二值模式纹理特征自适应加权

毕业设计说明书外文摘要

Title Realization of automatic face recognition system base on weighted LBP

Abstract Automatic face recognition has become a hot research topic in the field of artificial intelligence and pattern recognition。It has been widely used in many fields such as information security, self-service, electronic commerce and so on。Local binary pattern is a kind of operator that describes the local texture features of the image, which has the advantages of fast operation speed, gray scale invariance and rotation invariance。This topic implements an automatic face recognition system based on weighted LBP, with the aid of the tool OpenCV to achieve the process of face recognition。The interface is designed using MFC。Firstly, face is detected in the image。The face region is pided into a number of local areas and LBP texture features will be extracted from each local area。Then LBP texture features will be transformed into statistical histogram。By means of the Chi square test of the histogram, we can calculate the similarity between the histogram。With the weight of different local area,we can find the best matching object of the image to be recognized, and the face recognition task can be completed。

Keywords: Face recognition， local binary pattern, texture features, adaptive weight

1 引言 3

1。1 人脸识别概述 3

1。2 人脸识别技术国内外现状 4

1。3 本文的组织结构 5

2 加权局部二值模式综述 6

2。1 LBP算子概述 6

2。2 LBP算子的改进和变型 7

2。3 权重的计算 9

3 基于加权LBP人脸识别系统的设计与实现 11

3。1 实验环境 11

3。2 人脸数据库 12

3。3 总体设计 14

3。4 主要系统模块设计与实现 18

4 性能测试 27

4。1 性能评估指标 27

4。2 实验结果及分析 27

5 结论 32

5。1 论文总结 32

5。2 进一步工作展望基于加权LBP的自动人脸识别系统实现:http://www.youerw.com/jisuanji/lunwen_97160.html