Underwater Image Enhancement using Residual Injected Homomorphic Filtering and Wavelet Fusion.

View at publisher Summarise with VIRAT
Authors: Reena Mary George, S. Vishnukumar, Abel John, R Rajkumar
Cited by: 0
References: 26

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Underwater images are deformed due to wavelength dependent absorption of light and also due to scattering of light. This causes distortion of color, blurring and the fine details within images is lost. To correct this a frequency aware multistage underwater image enhancement framework is proposed that leverages color correction, complementary contrast enhancement, controlled frequency-domain detail injection, multiscale fusion and spatial refinement. To eliminate the color cast from the input image, a LAB-based color correction process is performed first that restores the colors and retains the original luminance profile of the image. The color corrected image then undergoes Rayleigh-based contrast enhancement to improve the overall visibility of the images with low-contrast underwater regions. The frequency-domain homomorphic filter is applied concurrently to enhance the reflectance components and to suppress the effect of nonuniform illumination. A residual frequency injection technique is then employed which will allow only selective high-frequency information to be incorporated into the spatially enhanced image. This avoids excessive amplification of frequency artifacts. The Rayleigh-enhanced image and the residual-injected frequency image are fused together using multiscale wavelet decomposition to obtain a desired balance between contrast stability and detail preservation. As the post processing step unsharp masking is applied to further enhance edge clarity and local details. Experimental results on benchmark datasets demonstrate that the proposed method achieves consistent improvements in visual quality and non-reference image quality metrics, such as UIQM and UCIQE.

Fields of Study & Subjects

Image Enhancement Techniques Underwater Vehicles and Communication Systems Image and Video Quality Assessment

Publication details

Journal / source
International Conference on Advances in Intelligent Computing and Applications (AICAPS)
Pages
1–7
Publication date
11 February 2026
Language
EN