提出了融合小波和贝叶斯的人脸识别方法。对原始图像采用小波分解后,原始图像被分解到不同的频带上。利用小波理论分析可知,在每一级分解中,低频子图像包含了原始图像的主要描述信息,而其他高频子图像包含的信息较少,对模式分类的作用也较小,所以可忽略不计。该算法首先对图像进行二级小波分解,其次对得到的每幅低频子图进行贝叶斯人脸识别。在FERET 人脸库的子集上对识别算法进行了测试和比较。实验表明,与传统的方法相比较,该方法降低了运算量,提高了识别率。关键词:人脸识别;小波变换;贝叶斯方法Abstract: A new algorithm for face recognition based on wavelet transform and Bayesian is proposed. The original image is decomposed into low frequency and high frequency sub-band images by applying wavelet transform. According to the wavelet theory, the low frequency image is the smoothed version of the original image and the best approximation to the original image with lower-dimensional space. It also contains main energy content within the original image. But the other high frequency sub-band images contain less energy content and are almost useless to pattern discrimination. Then, Bayesian approach is used to thelow-frequency sub-band. Its efficiency and superiority are clarified by comparative experiment on a subset of FERET face data.
猜您喜欢
评论