Abstarct: Currently, due to the spread and high volume of digital information, image compression is presented as a basic requirement in the archive and sending systems. Therefore, a lot of methods have been established to increase the compression rate of image. However, increasing compression rate of image and maintain of the retrieved image is still regarded as a necessity. Each of the available compression method attempts to identify the redundancy and consequently to compress it baxsed on one or more features. Accordingly, each of the image compression algorithms doesn’t present the same compression rate for different images. Compression rates of these algorithms may depend on some factors such as the variety of the grey levels, smoothness of the image and the uniformity of the adjacent pixels in the image. Accordingly, some solutions are presented in this study which is applied on the image as a pre-processing stage so that they reinforce the effective parameters in the image toward the compression. The firs processing is lossless and formed in two ways. In the first step image is divided into non-overlapping blocks and the blocks are arranged baxsed on the coefficient increase (decrease) of DC. In the second step, eight reversible geometric transforms are shaped on the resulted image of the first step. Among the resulted images, an image is selected which leads to the best result for compression. This process reduces the scope of the changes in the results of the JPEG calculations and improves the compression rate of it. In the second method, the Grey levels that are quantitatively closer are categorized in one group. Then, a grey level is considered as a representative for each group. The grey level value of each pixel of the original image is replaced with the related representative value of its group. That is why this value has been chosen so that the minimum destruction of the image occurs. In the next stage, the histogram of the pre-processed image is packed. This pre-process as well as the preservation of the image quality, improves the compression rate of lossless JPGE. Third method is same with second method with this difference that image is blocked in fixed sizes and value of the parameter is selected baxsed on the human vision system for each block. This pre-processing improves the compression rate of JPEG2000 and Lossless JPEG algorithms. In the fourth method, using Genetic algorithm, a suitable color space is created for compressing each image separately. This pre-process on compression algorithm has improved performance of CIC algorithm.