Modified Heuristic Requantization Algorithm for JPEG Images

Several techniques have been developed for recompressing the JPEG image in order to achieve low bit rate and to maintain image quality. In this paper, we have modified the existing heuristic algorithm for requantizing JPEG images. The resulting images are generally smaller in size and have improved perceptual image quality over “heuristic” and “direct” (decompressing the JPEG image and then compressing it again with a larger step size)requantization approaches. We have compared the recompression results for standard benchmark images against image quality measures such as bits per pixel, compression ratio, image size and PSNR.