Image
Image Compression and Retrieval Techniques for Multi-Application Smart Card
1L.M. PALANIVELU and 2Dr. P. VIJAYA KUMAR, M.E. PhD
1 Associate Professor - CSE department, Velalar College of Engineering & Technology, Erode, India
E-mail: [email protected]
2 Department of EEE, Karpagam College of Engineering, Coimbatore, India
Abstract: Smart cards future is based on its ability to contain data in its memory and perform operations like comparing data with external databases, computing of digital signatures from other data and including multiple applications. Smart cards should be able to store images like photos, biometric and medical images required for varied applications. Thus, images in the smart card are compressed. This study focuses on the image retrieval problem using compressed images and the impact of compression in the classification accuracy. In this paper, we investigate lossy compression techniques required for various images in multi-application smart card framework and classification accuracy for retrieving compressed image. The images are decomposed using Symlet wavelets. In this work, we propose to compress images including photo of the smart card holder, biometric information and medical images using Embedded Zero trees of Wavelets (EZW) algorithm. Features from the image are extracted using Gabor filters. The classification accuracy for retrieval of uncompressed images and various compressed images is evaluated using Naïve Bayes.
[L.M. PALANIVELU and P. VIJAYA KUMAR. Image Compression and Retrieval Techniques for Multi-Application Smart Card. Life Sci J 2013;10(4s):531-541] (ISSN: 1097-8135). http://www.lifesciencesite.com. 81
Keywords: Smart Card, Image Compression, Symlet Wavelet, Gabor filters, Embedded Zero trees of Wavelets (EZW).