The biometric identification system using human palm has been developed and presented in this paper. Palmprints have been retrieved based on Principal Component Analysis (PCA) with Gaussian Normalization (GN) and Self-Organizing feature Map (SOM). An integration of PCA, GN and SOM is proposed, where the coefficients obtained by PCA is normalized by GN for global feature representation is considered as input features for SOM. The trained SOM can be used as a retrieval engine to identify similar palmprint images for individual identification. The proposed system has been tested with different images. Experimental tests have proved that the proposed approach is not only robust but also quite efficient.