Compression of Fingerprint Using SVD in Sparse Respresentation.

Compression is known as reducing the memory size of an image. In the process of compression, there will be a loss of data then this type of compression is called lossy compression. If there is no loss of data, then this type of compression is called lossless compression. This new fingerprint compression algorithm is based on Singular Value Decomposition in sparse representation. In the algorithm, firstly the uploaded image is divided into patches then secondly dictionary is created for each fingerprint image patch. These patches are compressed using SVD in sparse representation. Then the image is reconstructed. This is a Lossless compression. The experiment demonstrates that our algorithm is efficient compared with several competing compression techniques (JPEG, JPEG 2000, and WSQ), especially at high compression ratios. Conclusions: Sparse representation is used for compression. JPEG has bad performance at low-bi rates and this is the more sophisticated algorithm. It provides a high Peak Signal to Noise Ratio and a high compression ratio. In this paper, a Fingerprint image is compressed using Singular Value Decomposition(SVD) in Sparse representation. The uploaded image occupies more storage size and requires more bandwidth and time to transfer. In this algorithm, Upload a fingerprint image from system memory into the application. This image is divided into patches. Dictionary is created for each patch of the image. SVD is used for the compression of the image. Reconstruction of the compressed image takes place. The resultant image after compression will have less storage size with the same quality. Bandwidth and time taken for the image transfer are less now. A graph is represented comparing the storage size of the uploaded image and the compressed image. [ABSTRACT FROM AUTHOR]