Studies on Underwater Image Processing Using Artificial Intelligence Technologies

Sugunapriya

doi:10.1109/ACCESS.2024.3524593

Return

Studies on Underwater Image Processing Using Artificial Intelligence Technologies

2025-01-01

Sugunapriya

S Markkandan *

DOI：10.1109/ACCESS.2024.3524593

Original

Original request for help

Save

Abstract

En 中文

Underwater image processing is a dynamic field that has garnered increasing interest due to its critical applications in marine biology, geological explorations and military reconnaissance. This paper presents a comprehensive survey of the methodologies employed in the enhancement and restoration of underwater imaging with a specific focus on the challenges posed by aquatic medium such as light absorption, scattering and color distortion. The survey reviews a range of techniques from traditional histogram equalization and white balancing methods to cutting edge AI approaches, including CNNs and GANs. Through examination of various datasets and quality metrics, we assess the performance of these methodologies in overcoming the inherent challenges of undersea imaging. Our study highlights the significant advances in AI driven underwater image processing technologies with a need for more resilient and flexible algorithms that can manage the intricacies of an undersea environment. The findings of this survey suggest promising directions for future research, particularly in the development of more sophisticated deep learning models that can further improve image quality and contribute to the underwater exploration and monitoring system.

Keywords:

Image color analysis

Surveys

Image enhancement

Image processing

Histograms

Image recognition

Deep learning

Object detection

Measurement

Videos

Underwater image enhancement