Sparse Representation based Fingerprint Compression

Recognition of people by means of their biometric characteristics is very popular among the society. There are various biometric techniques including fingerprint recognition, face recognition and eye detection that are used for the privacy and security purposes in different applications. Among all these techniques, fingerprint recognition has gain more popularity for personal identification due to its unique structure. Large volumes of fingerprint are collected and stored every day in a wide range of applications. Therefore, compression of these data become very important under specific circumstances due to this large amount of data involved. The requirement for this huge amount of memory is need to be reduced for the purpose of efficiency. Fingerprint image compression is a key technique to solve this problem. The various techniques have been proposed in the past for efficient fingerprint compression. This paper extends the sparse representation approach of fingerprint compression by performing enhancement with DWT along with thinning and binarization at pre-processing stage. The proposed approach can work for both high and low intensity sensors.