Exploring Conventional Approaches for Color Image Denoising: A Comparative Study

Abstract

Image denoising plays a critical role in enhancing the quality of digital images by removing unwanted noise while preserving important image details. Among various noise types, color image denoising presents unique challenges due to the complex correlation between color channels. This paper explores conventional denoising approaches specifically tailored for color images, focusing on well-established techniques such as median filtering, Gaussian smoothing, bilateral filtering, Non-Local Means (NLM), and wavelet-based denoising. Each method is analyzed for its effectiveness in suppressing noise while maintaining image integrity. We perform a comparative study to evaluate the performance of these techniques across different noise models, including Gaussian, salt-and-pepper, and speckle noise. Objective metrics such as Peak Signal-to-Noise Ratio (PSNR) and Root Mean Square Error (RMSE) are used to assess image quality post-denoising. Our results highlight the strengths and limitations of each method, offering insights into which conventional approaches are most suitable for specific noise types and image content. This comparative analysis serves as a foundation for further research and development of advanced denoising techniques.