A wavelet-based low frequency prior for single-image dehazing

Abstract The ever-growing sector of computer vision and image processing demands real-time enhancement techniques to satisfactorily restore hazy images. Although dark channel prior is most notable for single image haze removal, its major drawback is its large processing time. In this chapter, we propose a time-efficient wavelet-based prior, namely low-frequency prior. Low-frequency prior assumes that the majority of haze is contained in the low-frequency components of a hazy image. Here, we have transformed the hazy image into the wavelet domain using discrete wavelet transform to segregate the low- and high-frequency components, and treat them accordingly. Only the spatial low-frequency components are subjected to dark channel prior dehazing. The obtained dehazed image with low contrast can then be subjected to the novel fuzzy contrast enhancement framework presented in this chapter. Qualitative and quantitative comparisons with other state-of-the-art methods prove the primacy of the proposed framework.