Fuzzy logic based vision enhancement using sigmoid function
Sriparna Banerjee, Sheli Sinha Chaudhuri, Sangita Roy · 2017
In this paper, the image haze-removal technique proposed by us is free from artifacts as well as it's execution time is also very less. The method proposed here can overcome the halo-artifacts occurring due to non-uniform transmission within patches by patch size regulation and is also capable of removing the blocking artifacts (especially in the sky region) and can restore the contrast of the image by using sigmoid function and fuzzy logic. Moreover, since we have taken a large number of hazy images in our database and have obtained satisfactory results by applying the algorithm on all the hazy images, hence we can state that our algorithm is not dependent on the features of the hazy images and can remove haze from all kinds of hazy images irrespective of its characteristics.