| 一种改进的快速人脸检测算法 |
| |
| 引用本文： | 王小兰,蔡灿辉,朱建清. 一种改进的快速人脸检测算法[J]. 微型机与应用, 2013, 32(15): 37-40 |
| |
| 作者姓名： | 王小兰 蔡灿辉 朱建清 |
| |
| 作者单位： | 1. 华侨大学信息科学与工程学院,福建厦门,361021
2. 中国科学院自动化研究所,北京,100190 |
| |
| 摘 要： | 针对AdaBoost人脸检测算法易受到纹理干扰而造成误检的情况,提出了一种结合自适应肤色验证的快速人脸检测算法CMFS。该算法首先用运动检测为前置处理减小人脸检测搜索范围,然后用AdaBoost算法检测出人脸候选区,最后根据人脸候选区的平均亮度自适应地选择是否启用肤色验证作为后处理以排除虚警。实验结果表明,提出的CMFS检测算法在保证检测率的同时,提高了检测速度,并大大降低了误检率。
|
| 关 键 词： | 人脸检测 肤色检测 多特征融合 |
An improved fast face detection algorithm |
| |
| Wang Xiaolan , Cai Canhui , Zhu Jianqing. An improved fast face detection algorithm[J]. Microcomputer & its Applications, 2013, 32(15): 37-40 |
| |
| Authors: | Wang Xiaolan Cai Canhui Zhu Jianqing |
| |
| Affiliation: | 1 .Institute of Information Science and Technology , Huaqiao University , Xiamen 361021 , China ; 2 .Institute of Automation , Chinese Academy of Sciences , Beijing 100190 , China) |
| |
| Abstract: | In this paper, a concatenate detection algorithm combining adaptive skin-color verification, which is named CMFS is proposed to reduce the false rate of face detection algorithm based on AdaBoost.CMFS uses motion detection to reduce the search region of face detection, then uses AdaBoost face detection algorithm to detect face candidates, finally according to the average brightness of each face candidate adaptively chooses skin -color verification as post -processing to exclude the false alarm.Experimental results have shown that the proposed CMFS algorithm can greatly improve the detection rate and reduce the false rate of face detection algorithm. |
| |
| Keywords: | face detection skin detection multi-featunes combination |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
