﻿ 基于Haar特征与残差网络实现人脸检测的对比 The Contrast between the Haar Feature and the Residual Network for Facial Recognition

Journal of Image and Signal Processing
Vol. 08  No. 02 ( 2019 ), Article ID: 29604 , 5 pages
10.12677/JISP.2019.82009

The Contrast between the Haar Feature and the Residual Network for Facial Recognition

Kang Rong, Mingxin Jiang*

School of Electronic Information Engineering, Huaiyin Institute of Technology, Huai’an Jiangsu

Received: Mar. 13th, 2019; accepted: Mar. 24th, 2019; published: Apr. 8th, 2019

ABSTRACT

Face detection is a kind of computer technology which can analyze and distinguish facial features by extracting them. This paper introduces two face detection algorithms based on Haar feature and residual network. The advantages and disadvantages of the two algorithms are analyzed by comparing the detection speed, the brightness of ambient lights and the integrity of human faces.

Keywords:Haar Feature, Residual Network, Face Detection

1. 前言

2. 基于Haar特征的人脸检测

2.1. 什么是Haar特征

Figure 1. Feature template

2.2. 基于Haar特征的检测原理概况

${h}_{J}\left(x\right)=\left\{\begin{array}{c}1\\ 0\end{array}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\begin{array}{c}\text{if}p*f\left(x\right)

2.4. 级联分类器

3. 基于残差网络的人脸检测

3.1. 什么是残差ResNet网络

3.2. 残差Resnet网络的结构

Figure 2. Residual structure

4. 实验结果及分析

Figure 3. Face detection effect map based on Haar feature

Figure 4. Face detection effect map based on residual network

5. 总结

The Contrast between the Haar Feature and the Residual Network for Facial Recognition[J]. 图像与信号处理, 2019, 08(02): 60-64. https://doi.org/10.12677/JISP.2019.82009

1. 1. 黄泉龙. 基于PCA的人脸识别研究[D]: [硕士学位论文]. 西安: 西安电子科技大学, 2012.

2. 2. 王启航. 基于深度卷积神经网络的人脸检测算法研究[D]: [硕士学位论文]. 杭州: 浙江理工大学, 2018.

3. 3. 张宁. 基于PCA算法的人脸识别研究[J]. 山西电子技术, 2009(2): 23-24.

4. NOTES

*通讯作者。