>> 自然科学版期刊 >> 2008年05期 >> 正文
基于Hough变换的人眼检测方法研究
供稿: 吴冰;陈步威;李富贵;杨文强 时间: 2019-04-26 次数:

作者:吴冰;陈步威;李富贵;杨文强

作者单位:河南理工大学电气工程与自动化学院

摘要:人脸检测在许多应用中都是重要的一个处理阶段,眼睛是一个在人脸检测中极为重要的人脸特征,因此一种精确可靠的定位眼睛的算法对许多实际的应用是十分重要的.提出一种基于Hough变换椭圆检测的人眼睛定位方法,先对人脸图象进行canny边缘提取,然后根据眼睛的生理特征,利用提取的眼睛边缘特征进行椭圆检测.实验证明该方法定位准确,能有效的抑制噪声,并且有很高的识别率.

基金:河南省教育厅自然科学研究计划项目(2008B510009);

关键词:Hough变换;椭圆检测;眼睛检测;

DOI:10.16186/j.cnki.1673-9787.2008.05.022

分类号:TP391.41

Research on detection of human eye based on Hough transfor m

Abstract:Face detection is an important step in many applications such as human face recognition, video conferencing, human computer interfacing etc.The eye is the most important facial feature in face detection.Thus, a reliable and fast method for locating the eyes precisely in an image is vital to many practical applications.A new method for detection and recognition of human eye is proposed in the paper, based on ellipse and Hough transform.At first, images of the human face edge are extracted based on canny method.Then, according to the physiological characteristics of the eye, ellipse detection is done from the eyes of extracting edge features.The results show that this method has positioning accuracy, effective noise suppression, and a high recognition rate.

最近更新