We cannot operate the whole image directly in some applications like image recognition or image compression because it is inefficient method. Therefore, before recognition or compression, many image segmentation algorithms were suggested to segment an image. Image segmentation operation is to rate or cluster an image into many regions depending on the feature of image. A hybrid image compression algorithm is suggested in this paper which segments the image into background and foreground parts and after that compresses the foreground image by using DCT technique. The foreground image is provided more importance than the background region. The color based segmentation method is used to segment image. The observed parameters are compression ratio (CR), mean square error (MSE), and peak signal to noise ratio (PSNR). The goal is to maximize the CR while preserving images' information. The foreground image has a good quality in this the proposed hybrid method.