A no-reference image quality index for contrast and Sharpness measurement

摘要

Image quality evaluation plays a very important role in quantifying the quality of an image. Most of the indices already proposed can quantify only single types of distortions and they are also not very well correlated with Human Visual System (HVS). Contrast and Sharpness are the two parameters on the basis of which the proposed quality index will evaluate the performance of image enhancement algorithms. The degree of contrast will be evaluated by considering the difference in the average gray level values in its foreground to that of background. The calculated statistical parameters for foreground and background regions are mathematically combined using the Logarithmic Image Processing (LIP) operators to ensure processing of images from HVS point of view. Another parameter, sharpness will be calculated using wavelet decomposition method employing Haar transform. Finally, both the parameters will mathematically combine to assess the quality of enhanced image. Simulation results illustrate the precision and efficiency of the proposed quality index in comparison to previously proposed measure for assessing the image enhancement.