Hybrid Image Denoising using Adaptive Conductance Function and Bi-dimensional Empirical Mode Decomposition

In digital imaging, effective noise reduction is essential for maintaining image clarity, especially in fields such as medical imaging, remote sensing, and computer vision. Traditional denoising techniques often involve a trade-off between noise suppression and detail preservation, which can result in loss of essential image features. This paper presents a novel hybrid denoising approach that combines an Adaptive Conductance Function (ACF) with Bi-dimensional Empirical Mode Decomposition (BEMD). The proposed method leverages the adaptive properties of ACF to dynamically reduce noise based on local image gradients, preserving edges and fine details. The BEMD component decomposes the image across multiple frequency scales, allowing selective noise reduction on high-frequency components without affecting the underlying structural information. Experimental evaluations on synthetic and real-world datasets, including images with Gaussian and salt-and-pepper noise, demonstrate significant improvements in Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) compared to traditional methods. This hybrid method shows strong potential for applications requiring high-fidelity image denoising and detail preservation.