
Edge Aware Turbidity Restoration of Single Shallow Coastal Water Image
Author(s) -
S. Mary Cecilia,
Suriya Murugan
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1911/1/012021
Subject(s) - image restoration , artificial intelligence , computer vision , computer science , enhanced data rates for gsm evolution , contrast (vision) , turbidity , underwater , channel (broadcasting) , filter (signal processing) , noise (video) , waves and shallow water , image (mathematics) , image processing , environmental science , geology , computer network , oceanography
The blur and low contrast of underwater images are indicative of the high noise, intense scattering and ultimately the low quality of such images. Image Enhancement and Restoration are imperative pre processing steps for single images from shallow coastal areas. Pre processing of underwater images is hence a prerequisite to processes like classification, object detection and computer vision. The paper presents an effective edge aware restoration models that goals turbid water images. The dark channel based restoration method with a rolling guidance filter gives a more edge aware restored and denoised version of the heavily blurred, low contrast shallow coastal image. This is evident from the subjective and objective projections. Moreover the edge preserving nature projects 13.05% and 5.53% more than UHP (which is in close) in terms of number of edges and UCIQE scores in comparisons with promising algorithms over the decade.