Analysis of relationship between size of image resolution and recognition performance

Palmprint recognition for identification provides a new scheme for information security. In order to begin with practice and reduce restrictions on the image acquisition, the paper presents independent component analysis full of the better performances to extract the features of the palmprint images on the basis of wavelet transform. It uses the simple nearest neighbor classifier to clarify. It studied the relationship between the size of the palmprint resolution and the recognition performance. The images of different resolution were tested on the basis of UST palmprint image database. From the experimental results, the recognition result is concerned with the studied palmprint region and its size when the recognition rate is guaranteed. The least resolution may be reduced to three sixty-fourths of the original image. So it increases the system speed.