Wave atoms based compression method for fingerprint images

This paper proposes a new fingerprint image compression approach where the quality of the decoded image is perceptually controlled using wave atom transform. A comparative study of different transforms shows that wave atom transform is more appropriate than wavelets for fingerprint image compression. A new image quality metric based on the same transform is proposed to control the compression performance. Some properties of the human visual system are exploited in the design of the proposed image quality metric. Simulations show that this metric correlates well with the subjective human judgment. Based on these interesting results a compression method with a perceptual distortion control mechanism specific to fingerprint images is developed. Finally, the performance of the proposed compression scheme is evaluated in terms of fingerprint recognition rate. The experimental results clearly show that the proposed strategy offers better results than some traditional compression methods.

► A comparative study of geometric wavelets for compression is performed.
► The Wave Atom (WA) transform is found suitable for fingreprint compression.
► An adaptive scalar quantization derived from JPEG2000 is used.
► The distortion is controlled using a WA-based Full Reference image quality metric.
► The overall system is evaluated by means of fingerprint recognition system.