Conclusion

The greatest deficiency of the Y-JPEG scheme is false contours in low-intensity regions due to insufficient quantization levels. On the other hand, in the high intensity regions the Y-JPEG scheme preserves mean levels better than the traditional JPEG scheme.

Since the intensity levels can not be converted back to framebuffer values until after decompression, this scheme can not be integrated with the existing format to become an option for bright images. The benefits of the Y-JPEG compression is not very obvious. (It depends also on monitor setup: The brightness difference is observable on some but not all of the computers in Sweet Hall, for example.) Meanwhile, the artifacts in the low intensity regions are almost always noticeable and objectionable.

If you have questions or comments, please e-mail Alan Tseng / alant@stanford.edu.