2018/12/04 by Sanchayan Santra, Ranjan Mondal, Santra, Sanchayan +7
Computer Science · Engineering · #Advanced Image Fusion Techniques #Advanced Image Processing Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Enhancement Techniques #Video Surveillance and Tracking Methods
paper · pdf · doi:10.48550/arxiv.1812.01273
openalex publication_date 2018/12/04 · openalex created_date 2022/08/01 · openalex updated_date 2026/07/28
Haze limits the visibility of outdoor images, due to the existence of fog,\nsmoke and dust in the atmosphere. Image dehazing methods try to recover\nhaze-free image by removing the effect of haze from a given input image. In\nthis paper, we present an end to end system, which takes a hazy image as its\ninput and returns a dehazed image. The proposed method learns the mapping\nbetween a hazy image and its corresponding transmittance map and the\nenvironmental illumination, by using a multi-scale Convolutional Neural\nNetwork. Although most of the time haze appears grayish in color, its color may\nvary depending on the color of the environmental illumination. Very few of the\nexisting image dehazing methods have laid stress on its accurate estimation.\nBut the color of the dehazed image and the estimated transmittance depends on\nthe environmental illumination. Our proposed method exploits the relationship\nbetween the transmittance values and the environmental illumination as per the\nhaze imaging model and estimates both of them. Qualitative and quantitative\nevaluations show, the estimates are accurate enough.\n