Approximate solution to optimized contrast enhancement for hazy image
Jin Hwan Kim, Chang‐Su Kim · 2014
An approximate solution to single image dehazing algorithm is proposed in this work. We improve the performance of our previous dehazing algorithm, which optimally estimates transmission values, by approximating a cost function. The cost function consists of two terms: contrast cost and information loss cost. The information loss cost is nonlinear, which cannot be converted to a quadratic form. To address this issue, the proposed algorithm applies the logistic function and replaces the information loss term to the logistic form, thereby achieving complexity reduction. Experimental results demonstrate that the proposed algorithm can remove haze efficiently and reconstruct details in original scenes clearly.