用于身份鉴别的掌纹识别为信息安全提供了一种新的方案。为减少对图像采集的限制,提出在小波变换的基础上,利用高低帽变换寻找低频子图像中的灰度槽,获取对比度增强的图像。把此子图像所有的变换值组合起来作为图像的特征向量用于识别;运用UST掌纹图像库,对本算法进行了测试。从实验的结果看,此方法解决了在线掌纹图像低对比度问题,图像识别率得以提高,能够满足对采集图像无过多要求的认证系统的使用。
Palmprint recognition for identification provides a new scheme for information security.In order to reduce restrictions on the image acquisition,the paper presented high and low hat transform method to find the grey tanks in the low frequency sub-image and got the contrast-enhanced image on the basis of wavelet transform.Grouped all the transform values in the sub-image into a feature vector for recognition.The method was tested on the basis of UST palmprint image database.From the experimental results,the method solved low contrast about on-line palmprint images so that increased the recognition rate.It proves that the method can satisfy the uses for authentication systems without excessive demands for collection images.