Single Image Visibility Restoration Using Dark Channel Prior and Fuzzy Logic

Symptomatic vitreous opacity image enhancement using linear and nonlinear transformations

“…HLIPSCS [31] overall image does not highlight local details. RCEA [32] correction process introduces undesired white and gray The SIVR [33] enhanced image is whitish overall and the enhancement information is not clear. SVR [34] corrects the contrast to some extent, but the overall image visibility is very low.…”

“…The halo phenomenon is manifested in the image by the appearance of edges with lower brightness at the edge parts of the highlighted regions. The SIVR [33] method is able to represent the detailed information of the image well, but the visibility of the image is low. The SVR [34] method does not significantly make enhancement to the image and cannot overcome the effect of the artifacts by the post-processing.…”

A Novel Image Dehazing and Assessment Method

“…. Estimation of A described in original DCP method is used in [21][22][23][24][25][26][27], [10] and many other methods. In this method, A is estimated by first selecting the indices the top 0.1% pixels of dark channel and then using these indices the maximum intensity pixel in the hazy image is selected for A.…”

“…Since different computer vision algorithms assumes that the input image is scene radiance [17]. Degraded image creates the hindrance in object detection, motion tracking, satellite imagery, aerial photography [9], autonomous driving and face recognition in CCTV security cameras [10]. This makes the task of haze removal to be inevitable before application of outdoor computer vision algorithms.…”