{
"title": "Fingerprint Compression Using Contourlet Transform and Multistage Vector Quantization",
"authors": "S. Esakkirajan, T. Veerakumar, V. Senthil Murugan, R. Sudhakar",
"country": null,
"institution": null,
"volume": "3",
"journal": "International Journal of Computer, Electrical, Automation, Control and Information Engineering",
"pagesStart": 823,
"pagesEnd": 831,
"ISSN": "1307-6892",
"URL": "http:\/\/waset.org\/publications\/7927",
"abstract": "This paper presents a new fingerprint coding technique\r\nbased on contourlet transform and multistage vector quantization.\r\nWavelets have shown their ability in representing natural images that\r\ncontain smooth areas separated with edges. However, wavelets\r\ncannot efficiently take advantage of the fact that the edges usually\r\nfound in fingerprints are smooth curves. This issue is addressed by\r\ndirectional transforms, known as contourlets, which have the\r\nproperty of preserving edges. The contourlet transform is a new\r\nextension to the wavelet transform in two dimensions using\r\nnonseparable and directional filter banks. The computation and\r\nstorage requirements are the major difficulty in implementing a\r\nvector quantizer. In the full-search algorithm, the computation and\r\nstorage complexity is an exponential function of the number of bits\r\nused in quantizing each frame of spectral information. The storage\r\nrequirement in multistage vector quantization is less when compared\r\nto full search vector quantization. The coefficients of contourlet\r\ntransform are quantized by multistage vector quantization. The\r\nquantized coefficients are encoded by Huffman coding. The results\r\nobtained are tabulated and compared with the existing wavelet based\r\nones.",
"references": null,
"publisher": "World Academy of Science, Engineering and Technology",
"index": "International Science Index 3, 2007"
}