학술논문
FEUWNet: A Fast and Effective Baseline for Underwater Image Enhancement
Document Type
Conference
Author
Source
2023 IEEE International Conference on Mechatronics and Automation (ICMA) Mechatronics and Automation (ICMA), 2023 IEEE International Conference on. :19-24 Aug, 2023
Subject
Language
ISSN
2152-744X
Abstract
Underwater optical images are one of the important media for humans to explore the oceans, but it is difficult to directly obtain high-quality underwater images because of the unique physical and chemical properties of underwater, most underwater images exhibit disadvantages such as color decay, low contrast, and blurred details. In order to solve the above problems, this paper designed a fast and effective baseline for underwater image enhancement called FEUWNet. FEUWNet uses a plug-and-play module consists with downsampling encoding block and up-sampling decoding block to balance enhancement performance and computational speed. After experimental comparison, FEUWNet has achieved good results in underwater image enhancement.