Fingerprints are today the most widely used biometric features for personal identification. With the increasing usage of biometric systems the question arises naturally how to store and handle the acquired sensor data. In this context, the compression of these data may become imperative under certain circumstances due to the large amounts of data involved. In distributed biometric systems, the data acquisition stage is often dislocated from the feature extraction and matching stage. In such environment the sensor data have to be transferred via a network link to the respective location, often over wireless channels with low bandwidth and high latency. Therefore, a minimization of the amount of data to be transferred is highly desirable, which is achieved by compressing the data before transmission. When a typical fingerprint card is scanned at 500 dpi, with 8 bits/pixel, it results in about 10Mb of data. For 200 million cards, the total size of the digitized collection would be more than 2000 terabytes. Fingerprint images are routinely sent between law enforcement agencies. Overnight delivery of the actual card is too slow and risky, and sending 10 Mb of data through a 9600 baud modem takes about three hours.

Standard methods for still noise-free, image coding are applied in various domains. Spatial image coding technique is basic method as applied in the pixel-domain of the image. Fourier transform methods, is widely used in practice. Today the best known image compression techniques are the JPEG2000 and the SPIHT algorithms. The fractal transform was viewed in the signal processing community as a computationally expensive and limited technique that only works when the image exhibits a high degree of self-similarity. In this paper, a novel approach that reduces the computational complexity of the standard fractal transform. We propose a fast no-search fractal image coding algorithm based on a modified gray-level transform. To improve the possibility of successful matching for a range block and a domain block, we introduce a modified gray level transform with more transform parameters to encode the blocks. In this work we are using two gray-level transforms, one for the parent blocks and other for the child blocks to speed up the encoding time.