Open AccessMulti-scale fusion for underwater image enhancement using multi-layer perceptron
Author(s) -
M. Sudhakara,
M. Janaki Meena
Publication year - 2021
Publication title -
iaes international journal of artificial intelligence
Language(s) - English
Resource type - Journals
eISSN - 2252-8938
pISSN - 2089-4872
DOI - 10.11591/ijai.v10.i2.pp389-397
Subject(s) - computer science , underwater , benchmark (surveying) , artificial intelligence , transmission (telecommunications) , channel (broadcasting) , perceptron , image restoration , process (computing) , computer vision , image (mathematics) , pattern recognition (psychology) , image processing , artificial neural network , telecommunications , geology , oceanography , geodesy , operating system
Underwater image enhancement (UIE) is an imperative computer vision activity with many applications and different strategies proposed in recent years. Underwater images are firmly low in quality by a mixture of noise, wavelength dependency, and light attenuation. This paper depicts an effective strategy to improve the quality of degraded underwater images. Existing methods for dehazing in the literature considering dark channel prior utilize two separate phases for evaluating the transmission map (i.e., transmission estimation and transmission refinement). Accurate restoration is not possible with these methods and takes more computational time. A proposed three-step method is an imaging approach that does not need particular hardware or underwater conditions. First, we utilize the multi-layer perceptron (MLP) to comprehensively evaluate transmission maps by base channel, followed by contrast enhancement. Furthermore, a gamma-adjusted version of the MLP recovered image is derived. Finally, the multi-scale fusion method was applied to two attained images. The standardized weight is computed for the two images with three different weights in the fusion process. The quantitative results show that significantly our approach gives the better result with the difference of 0.536, 2.185, and 1.272 for PCQI, UCIQE, and UIQM metrics, respectively, on a single underwater image benchmark dataset. The qualitative results also give better results compared with the state-of-the-art techniques.