On the other
hand other metrics are in favor of JPEG by considering only a video
stream coming from a compression and decompression operation.
The results for the JPEG, the compression ratio, data rate, and PSNR
are reported in Table 3 by comparing results achieved by the proposed
technique. Moreover, overall JPEG results are reported in a
graphical form in Fig. 8, by showing the q and PSNR behaviors as
a function of BRP and JPEG quality. The Quality parameter is appointed
as counterpart of BRP in JPEG since the latter is undefined.
As it can be seen, in the ideal case of noise-free transmission, JPEG
outperforms the proposed compressor. In fact fair quality in the
perceived stream (e.g., PSNR ’ 35) corresponds to a compression
ratio double in JPEG than in our proposal. If we set the compression
ratio (i.e., q ’ 20), the quality of reconstruction obtained by JPEG
compressor is higher than in our case (35.23 dB for our compressor,
39.04 dB for JPEG).