Dehazed Image Enhancement by a Gamma Correction with Global Limits

Abstract

Single image haze removal has been an active research topic in the image enhancement or image restoration community. Most of researchers have put their effort to improve the dehazing performance through modifying algorithm itself. In this paper, we propose a post-processing scheme, which is a gamma correction, to enhance the visual quality of a dehazed image after some dehazing algorithm. It is well-known that the conventional gamma correction (CGC) severely suffers from the due distortion in many cases. Thus, it hinders the application of the CGC to dehazed image enhancement. By our observations, the problem is caused by the local limits, which are found in the RGB components separately. To relieve the problem, a gamma correction with global limits (GCGL) is introduced, where the global limits are obtained by the minimum and maximum in the RGB components. The proposed GCGL is justified and applied to dehazed image enhancement. The simulation results indicate that the visual quality of dehazed images have been generally improved in the given examples.