Colour Image Compression Through Hybrid Approach

Abstract

A colour image compression is one of the challenging tasks in the field of multimedia. In this paper an effort is made to compress the colour image using a hybrid combination of DCT, SVD and RLE. In this method, the red component, the green component and the blue component of the image are considered individually. At first, the red component of the image is made to undergo DCT and its DC-coefficient is stored separately. Then the transformed matrix is truncated using a threshold value. Then, it is decomposed using SVD. This gives decomposed matrices. Then, these decomposed matrices are truncated using a suitable threshold value. After that, the decomposed matrices are multiplied. The resultant matrix is again truncated using a threshold value. Since in the obtained matrix, majority of the elements are zero, it is converted into a sparse matrix form. In the sparse matrix notation, to reduce the redundancy, again run length coding is applied. Then the compressed form of the red component is obtained. Similarly, the green component and the blue component are also compressed. Then the performance parameters such as Mean Square Error, Peak Signal to Noise Ratio, Compression Ratio, Structural Similarity Index Measure, and Quality Index are evaluated.