Re-encoding JPEG images for smart phone applications

Capturing, storing and browsing images from smart phones is a task performed by any user every day. ISO/IEC JPEG is the commonly used standard for lossy compression of digital images. During the more than twenty years from JPEG standardization, novel image coding algorithms have been developed providing superior compression capabilities. In particular, the recent JPEG XR, providing high compression ratios and low computational complexity is an interesting candidate for the development of novel applications for smart phones. In this paper, a comparison between the JPEG XR and JPEG 2000 as tools for re-encoding JPEG images taken by smart phone camera is presented. Experimental tests on image data set are performed in order to collect numerical evidence of the distortion introduced by the re-encoding process.