Quality is a very important parameter for all objects and their functionalities.\nIn image-based object recognition, image quality is a prime criterion. For authentic\nimage quality evaluation, ground truth is required. But in practice, it\nis very difficult to find the ground truth. Usually, image quality is being assessed\nby full reference metrics, like MSE (Mean Square Error) and PSNR\n(Peak Signal to Noise Ratio). In contrast to MSE and PSNR, recently, two\nmore full reference metrics SSIM (Structured Similarity Indexing Method)\nand FSIM (Feature Similarity Indexing Method) are developed with a view to\ncompare the structural and feature similarity measures between restored and\noriginal objects on the basis of perception. This paper is mainly stressed on\ncomparing different image quality metrics to give a comprehensive view. Experimentation\nwith these metrics using benchmark images is performed\nthrough denoising for different noise concentrations. All metrics have given\nconsistent results. However, from representation perspective, SSIM and FSIM\nare normalized, but MSE and PSNR are not; and from semantic perspective,\nMSE and PSNR are giving only absolute error; on the other hand, SSIM and\nPSNR are giving perception and saliency-based error. So, SSIM and FSIM can\nbe treated more understandable than the MSE and PSNR.
Loading....