source code 位置:
samples/cpp/tutorial_code/gpu/gpu-basics-similarity/gpu-basics-similarity
The PSNR (peak SNR) returns a float number, that if the two inputs are similar between 30 and 50 (higher is better).
The SSIM returns the MSSIM of the images. This is too a float number between zero and one (higher is better), however we have one for each channel. Therefore, we return a Scalar OpenCV data structure.
MSSIM(Mean Structure Similitary Index) is a measure of distortion static images. It s comparing distorted image with reference image and as the result return value between 0 and 1, value 1 is only reachable in the case of two identical sets of data.
see also: http://opencv.itseez.com/doc/tutorials/gpu/gpu-basics-similarity/gpu-basics-similarity.html