Key-byte algorithm for image compression

Image size is increasing every day, according to their accuracy and resolution; where the memory needed to store these images become larger with time, Image compression algorithms are the utilities used to utilize the memory resources; this is done by reducing the space needed to store these images, natural images are the best images to use in testing the compression images; since the adjacent pixels are close together but not 100% equal. The paper presents a compression algorithm that constrained on the compression factor, where it tries to get the highest compression factor with an accepted visual contribution to the reconstructed image, the algorithm works in two stages the first one called the Averaging technique, which compresses the image by converting the image into a metric of averages, the second stage is called the Key-Byte technique, in this stage the output of the first stage divided into blocks and extract three new values for each block those are Average, MinMax, and Key-Byte values, this means three new metrics one holds the averages of the blocks, another one to holds the difference between the maximum and minimum values in each block, the third holds 0 or 1 according to the relation between the original average and the new averages.