On effective palmprint retrieval for personal identification

This paper presents a novel retrieval method for effective search of palmprints based on Principal Component Analysis (PCA) and Self-Organizing Feature Map (SOM). To reduce search space and speed up the query processing, an integration of PCA and SOM is proposed, where the coefficients obtained by PCA for global feature representation is considered as input features of SOM. The trained SOM can be used as a retrieval engine to identify similar palmprint images with respect to the query palmprint image for personal identification.