摘要
文章提出了一种基于彩色图像的24位面部识别方法,图像处理的主要部分在软件中占据非常重要的位置,图像质量直接影响位置确定和识别精度。通过背光补偿,高斯平滑和二进制进行判别。在判别之前,经过互补光对图像施行处置,而后通过肤色得到大致的人脸。最后,依照面部固有眼睛的对称性断定面部,提升了定位和判别精度。最终该系统应用与图书馆人脸识别,并有很好的效果。
In this paper,a 24-bit facial recognition method based on color image is proposed.the main part of image processing occupies a very important position in the software,and the image quality directly affects the location determination and recognition accuracy.It is distinguished by backlight compensation,Gaussian smoothing and binary.Before discrimination,the image is processed by complementary light,and the approximate human face is obtained by skin color.Then,the face is determined according to the symmetry of the inherent eyes of the face,and the accuracy of location and discrimination is improved.Finally,the system is applied to library face recognition,which has a good effect.
出处
《科技创新与应用》
2021年第7期86-88,共3页
Technology Innovation and Application
