Residual spatial entropy-based image contrast enhancement and gradient-based relative contrast measurement
Turgay Çelik, Heng-Chao Li · Journal of Modern Optics · 2016
This paper proposes a novel algorithm to achieve both global and local contrast enhancement using a new definition of residual spatial entropy of gray levels of an image. Residual spatial entropy is utilized to assign a weight to each gray level which is further used in mapping of input gray levels to output gray levels to achieve global contrast enhancement. A non-linear mapping is applied on transform-domain coefficients of image enhanced globally to perform local contrast enhancement. The algorithm allows to control levels of perceived global and local contrast. New definitions of full-reference relative contrast measures are also introduced. Experimental results show that proposed algorithm produces better or comparable contrast-enhanced images than several state-of-the-art algorithms.