A New Approach to Quantifying the Perceived Contrast of Complex Images


Extensive use of video content in various applications requires a solution to the problem of no-reference quantifying perceived contrast of an image. However, known methods of measuring integral contrast in an image have several disadvantages that significantly limit their use to quantify perceived contrast. In this paper, the problem of improving the accuracy of quantifying perceived contrast for complex multi-element images was considered. A new approach to no-reference quantifying the perceived contrast of the image was proposed. The proposed approach is based on the measurement of perceived contrast in the a reference image with a limited dynamic range, for which Weber-Fechner's law is valid, with subsequent correction of the measurement result taking into accounts the mean for the ratios between the contrast values of the pairs of objects in the reference and the source image. Based on this approach, a new technique for quantifying the perceived image contrast was proposed. Possible implementations of this technique for complete and incomplete integral contrast using weighted and relative contrast kernels proposed. Research confirms the effectiveness of the proposed approach to quantifying the perceived contrast of complex images.



Software And Hardware

• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL