Color Image Enhancement Based on Adaptive Nonlinear Curves of Luminance Features

Abstract

Abstract-This paper proposes an image-dependent color image enhancement method that uses adaptive luminance enhancement and color emphasis. It effectively enhances details of low-light regions while maintaining well-balanced luminance and color information. To compare the structure similarity and naturalness, we used the tone mapped image quality index (TMQI). The proposed method maintained better structure similarity in the enhanced image than did the space-variant luminance map (SVLM) method or the adaptive and integrated neighborhood dependent approach for nonlinear enhancement (AINDANE). The proposed method required the smallest computation time among the three algorithms. The proposed method can be easily implemented using the field-programmable gate array (FPGA), with low hardware resources and with better performance in terms of similarity.Index Terms-Luminance enhancement, color emphasis, tone mapped image quality index (TMQI), FPGA implementation

Improved Color Attenuation Prior for Single-Image Haze Removal

“…Additionally, a sophisticated adaptive tone remapping algorithm [31] also is used, as opposed to the simple tone remapping method used by Tarel et al [16]. As a result, this algorithm is a significant improvement upon the fast haze removal using median filters, and its hardware design presented by Ngo et al [35] provides an efficient means of handling large images in real time.…”

“…Tarel et al [16] employed a simple tone mapping that solely operates on the luminance channel to resolve this problem. We make use of an efficient adaptive tone remapping method, proposed by Cho et al [31], in this work,. Not only is the luminance enhanced, the color also is emphasized, according to the ratio of the enhanced luminance to the original luminance.…”

“…where EL denotes the enhanced luminance, L the input luminance, G L the luminance gain, W L the adaptive luminance weight, EC the emphasized color, C the input color, G C the color gain, and W C the adaptive color weight. Cho et al, [31] adopted an adaptive limit point, which is calculated according to the cumulative distribution function of the input luminance, to formulate the luminance gain G L as a nonlinear power function. The adaptive luminance weight W L is a simple linear function where the input luminance is the independent variable and W L itself is the dependent variable.…”

“…The adaptive color weight W C is the piecewise linear function consisting of three line segments, in which the first and the third lines are two thresholds. Readers can refer to Choe et al, [31] for a more detailed description and computational formulae.…”

“…We also propose using a simple low-pass filter to remove the background noise and using a simple linear weight to get rid of the color distortion. Furthermore, the commonly observed dynamic range reduction of the dehazed image is resolved by employing the adaptive tone remapping post-processing [31]. Finally, to lighten the CAP's intensive computation, the efficient quad-decomposition algorithm is utilized to estimate the atmospheric light and the simple modified hybrid median filter (mHMF) is used to refine the depth map.…”