Colored Image Compression Using Discrete Transformations for Sine and Cosine functions

Abstract

Image compression is one of the important aspects in image processing field. This paper discusses an idea for image compression, consists of the following steps: 1- Applying 1D-DCT on each row of an input colored image. 2- Applying 1D-DST on each column of matrix obtained from the first step.3- Quantization process is applied to increase number of zeros; this matrix is converted into one-dimensional array by scanning the matrix column by column.4- Using Run Length Encoding, to minimize the number of coefficients.5- Using Arithmetic coding procedure, to compress the minimized array. The decompression algorithm will reverse the compression steps used above. The decompressed image is compared with JPEG by Mean-Square-Error (MSE) and compression ratio as a measurement for the comparison between the conventional methods and our algorithm.Keywords: DCT, DST, Quantization, RLE, Arithmetic coding.