A Hybrid Adaptive Median-Gaussian Filter Effective Mixed-Noise Image Denoising and Feature Preservation

In this paper, Hybrid Adaptive Median-Gaussian Filter (HAMGF) designed to effectively eliminate different forms of noise, including salt-and-pepper and Gaussian noise, from images while preserving essential features without compromising perceptual quality. However, traditional denoising methods often focus on addressing either salt-and- pepper or Gaussian noise, resulting as suboptimal performance in mixed-noise environments. To overcome this limitation, the HAMGF employs an adaptive filtering strategy, applying hybrid adaptive median filtering technique especially overcome the problem of salt-and-pepper noise and an adaptive Gaussian filtering used for regions affected by particularly Gaussian noise. On the other hand, the filter dynamically adjusts its parameters based on local noise characteristics, ensuring effective denoising across a wide range of noise densities. Extensive experiments illustrate an excellent outcome of proposed HAMGF method in comparison with various existing state-of-the-art denoising techniques, including the Iterative- Adaptive-Median Filtering (IAMFA), Adaptive-Median Filtering (AMF), Weighted-Median-Filtering technique (WMF) and Hybrid Median-Filtering algorithm (HMF). Moreover, performance evaluations based on various standard metrics such as Structural Similarity-Index (SSIM), Peak-Signal-to- Noise-Ratio-(PSNR) and Correlation metrics reveal that HAMGF consistently outperforms the other methods, particularly where noise densities are very high. The outcomes highlight the effectiveness of adaptive hybrid filtering approaches in enhancing image restoration performance, and the proposed HAMGF filter demonstrates an excellent-level of versatility and resilience across different noise conditions.