A Review on Underwater Image Enhancement Using Deep Residual Framework

Authors

  • Prof. Anuja Phapale  Assistant Professor at Information Technology Department, AISSMS Institute of Information Technology, Pune, Maharashtra, India
  • Atal Deshmukh  B.E. Scholar, Information Technology Department, AISSMS Institute of Information Technology, Pune, Maharashtra, India
  • Keshav Katkar  B.E. Scholar, Information Technology Department, AISSMS Institute of Information Technology, Pune, Maharashtra, India
  • Onkar Karale  B.E. Scholar, Information Technology Department, AISSMS Institute of Information Technology, Pune, Maharashtra, India
  • Puja Kasture  B.E. Scholar, Information Technology Department, AISSMS Institute of Information Technology, Pune, Maharashtra, India

DOI:

https://doi.org/10.32628/IJSRST218314

Keywords:

Asynchronous training, edge difference loss, residual learning, underwater image enhancement

Abstract

There are various factors such as absorption, refraction & the phenomenon of scattering of light by particles suspended in water that are responsible for distorted colors, low contrast & blurred details of original underwater images. The traditional approaches include pre-processing the image using a descattering algorithm. The super-resolution (SR) method is applied. But this method has limitation that major part of the high frequency information is lost during descattering. This paper comes up with a solution for underwater image enhancement using a deep residual framework. Firstly, the generation of synthetic underwater images takes place for which cycle-consistent adversarial networks (CycleGAN) is employed. Further, these synthetic underwater images are used as training data for convolution neural network models. Secondly, the introduction of very-deep super-resolution reconstruction model to underwater resolution applications is carried out. Using this, the underwater Resnet model is proposed. It acts as a residual learning model for underwater image enhancement operations. Furthermore, the training mode & loss function are improved. Then, a multi-term loss function is formed which comprises of proposed edge difference loss & mean squared error loss. An asynchronous training mode is also being proposed that improves the performance of the multi-term loss function. Lastly, the discussion of the impact of batch normalization takes place. After a comparative analysis & underwater image enhancements, we can say that detailed enhancement performance & color correction of these proposed methods are much efficient & superior to that of previous traditional methods & deep learning models.

References

  1. G. P.Liu, "Underwater Image enhancement using deep residual framework," 2019.
  2. H.Qi, "Underwater Image enhancement using deep residual framework," 2019.
  3. C.Zhenge, "Underwater Image enhancement using deep residual framework," 2019.
  4. Z.Yu, "Underwater Image enhancement using deep residual framework," 2019.
  5. P. Luo, "Underwater Image super-resolution using deep residual multipliers," 2019.
  6. Y. Wang, "A deep CNN Methood for the Underwater image enhancement," 2017.
  7. J. Sattar, "Underwater image super-resolution with deep residual multipliers," 2019.
  8. H. H.Lu, "Underwater Image super-resolution by descattering and fusion," 2017.
  9. R.Hummel, "Image enhancement by histogram transformation,Compute Graph,Image Process," 2017.
  10. S.Corchs, "Underwater Image processing. State of the art of restoration and image enhancement methods," 2014.

Downloads

Published

2021-06-30

Issue

Section

Research Articles

How to Cite

[1]
Prof. Anuja Phapale, Atal Deshmukh, Keshav Katkar, Onkar Karale, Puja Kasture "A Review on Underwater Image Enhancement Using Deep Residual Framework" International Journal of Scientific Research in Science and Technology(IJSRST), Online ISSN : 2395-602X, Print ISSN : 2395-6011,Volume 8, Issue 3, pp.52-56, May-June-2021. Available at doi : https://doi.org/10.32628/IJSRST218314