Comparison of Compression Algorithms’ Impact on Iris Recognition Accuracy

The impact of using different lossy compression algorithms on the matching accuracy of iris recognition systems is investigated. In particular, we relate rate-distortion performance as measured in PSNR to the matching scores as obtained by a concrete recognition system. JPEG2000 and SPIHT are correctly predicted by PSNR to be well suited compression algorithms to be employed in iris recognition systems. Fractal compression is identified to be least suited for the use in the investigated recognition system, although PSNR suggests JPEG to deliver worse recognition results in the case of low bitrates. PRVQ compression performs surprisingly well given the third rank in PSNR performance, resulting in the best matching scores in one scenario. Overall, applying compression algorithms is found to increase FNMR but does not impact FMR. Consequently, compression does not decrease the security of iris recognition systems, but “only” reduces user convenience.