Recognition of palmprints using eigenpalms
Ajay Kumar, Helen C. Shen · 2003
This paper presents a new method for personal recognition using palmprints. This method uses, inkless, normalized palmprint images to generate eigenpalms. Every palmprint image is thus characterized by a feature vector, consisting of weights from eigenpalm images. The performance of proposed method using two measures i.e., minimum Euclidean distance and maximum similarity measure, is evaluated. The low-resolution images of 100 dpi, which dominantly capture the palmprint creases, have been used in the experiments. The experimental results show that this method achieves high recognition rate when the similarity criterion is used to recognize 300 inkless palmprint test images. 1.