An intelligent wireless channel corrupted image-denoising framework using symmetric convolution-based heuristic...

Area of Science:

• Signal Processing
• Image Processing
• Computer Vision

Background:

• Wireless image transmission is prone to noise, degrading image quality and hindering information extraction.
• Effective image denoising is crucial for accurate analysis and error correction in corrupted images.

Purpose of the Study:

• To develop an efficient image denoising approach for images corrupted during wireless transmission.
• To correct errors and mitigate channel degradation effects on image quality.

Main Methods:

• Images are decomposed using Adaptive Lifting Wavelet Transform (ALWT).
• A Symmetric Convolution-based Residual Attention Network (SC-RAN) is employed for residual image extraction.
• Parameters are optimized with the Hybrid Energy Golden Tortoise Beetle Optimizer (HEGTBO).

Main Results:

• The developed model achieves a Peak Signal-to-Noise Ratio (PSNR) of 31.69%.
• The proposed method demonstrates significant improvements in denoising corrupted images.

Conclusions:

• The integrated approach of ALWT, SC-RAN, and HEGTBO effectively enhances image quality.
• This research offers a robust solution for image denoising in wireless communication environments.