Paper Title:SVD Algorithm for Lossy Image Compression


Singular Value Decomposition as per it could be defined is the method of factorizing the real/complex matrix. These days the SVD algorithm tends to play a vital role in Image Processing. Its extending applications prove to be of a greater use. Not just image processing it also has equal importance in Statistics. It is the transformation of values in three different methods. In these different methods also the image compression can be done only through small set of values. The main objective of this methodology is to achieve this image compression using very less storage space. And also to preserve the most important set of values related to the image. The implementation is done through MATLAB. For example if m*n is a matrix SVD of A can be explained as the equation of A=UΣVT where, U is the product of m and m and orthogonal V is also the product of n and n and orthogonal and Σ is m*n diagonal matrix with all the positive set of values.

Keywords:Image Processing, Orthogonal, MATLAB, Diagonal matrix, Real and Complex matrix