摘要
首先对驾驶员图像采用同态滤波算法增强对比度,并用直方图均衡法进行预处理。再采用基于RGB、HIS等肤色模型的人脸检测方法与腐蚀、膨胀算法对图像进行优化,提取出脸部图像。最后通过水平与垂直投影检测眼部图像,并与眉毛区域相区分。
During the processing of extracted image of driver,the homomorphic filter was used to enhance the contrast of the image, and then the histogram equalization method was used during the preprocessing. Methods based on RGB and HIS skin color model,as well as corrosion and swelling algorithms were all used to get the image of drivers face. Finally, detection of the eye was achieved by vertical and horizontal projection, image of the eyebrow being distinguished at the same time.
出处
《邵阳学院学报（自然科学版）》
2017年第1期59-63,共5页
Journal of Shaoyang University：Natural Science Edition
关键词
直方图均衡
人脸识别
人眼识别
图像处理
histogram equalization
face recognition
recognition of human eyes
image processing
