IDE: Image Dehazing and Exposure Using an Enhanced Atmospheric Scattering Model

Abstract

Atmospheric scattering model (ASM) is one of the most widely used model to describe the imaging processing of hazy images. However, we found that ASM has an intrinsic limitation which leads to a dim effect in the recovered results. In this paper, by introducing a new parameter, i.e., light absorption coefficient, into ASM, an enhanced ASM (EASM) is attained, which can address the dim effect and better model outdoor hazy scenes. Relying on this EASM, a simple yet effective grayworld-assumption-based technique called IDE is then developed to enhance the visibility of hazy images. Experimental results show that IDE eliminates the dim effect and exhibits excellent dehazing performance. It is worth mentioning that IDE does not require any training process or extra information related to scene depth, which makes it very fast and robust. Moreover, the global stretch strategy used in IDE can effectively avoid some undesirable effects in recovery results, e.g., over-enhancement, over-saturation, and mist residue, etc. Comparison between the proposed IDE and other state-of-the-art techniques reveals the superiority of IDE in terms of both dehazing quality and efficiency over all the comparable techniques.

A Model-Driven Deep Dehazing Approach by Learning Deep Priors

“…We first evaluate our proximal dehaze-net on synthetic datasets and compare it with other single image dehazing methods. We select some representative single image dehazing methods, including DCP [4], CAP [8], NLD [9], GRM [10], IDE [11], MSCNN [15], DehazeNet [16], AODNet [17], DcGAN [20], GDN [22], DADN [27] and FDGAN [26]. Among these methods, DCP, CAP, NLD, GRM and IDE are traditional image processing methods based on image priors.…”

“…The traditional image dehazing methods [1]- [11] have investigated various haze-related image priors. Tan [2] assumes that the contrast of hazy images is lower than hazefree images and propose to maximize the contrast of hazy images under the MRF framework.…”

A Review of Remote Sensing Image Dehazing

“…In ASM, the first term on the right side, named Direct Attenuation, is used to describe the direct impact of scene reflection light caused by haze, which usually attenuates exponentially with the scene depth . The second term is called airight, which increases with the scene depth [ 23 ].…”

“…In other words, ASM considers that all scenes in the image are directly illuminated by the atmospheric light, while ignoring the influence of uneven illumination. To address the above problems, many useful methods [ 18 , 23 ] which optimize the robustness of the ASM are proposed. Although the dim effect is solved to a certain extent, uneven haze remains a challenge.…”

“…where x is the pixel coordinates, I is the observed intensity, J is the true radiance of the scene point imaged at x, A is the global atmospheric light, and t is the medium transmission. In ASM, the first term on the right side, named Direct Attenuation, is used to describe the direct impact of scene reflection light caused by haze, which usually attenuates exponentially with the scene depth d. The second term is called airight, which increases with the scene depth d [23].…”