A New Image Denoising Method Based On Region Growing Segmentation

Duman E., ERDEM O. A.

25th Signal Processing and Communications Applications Conference (SIU), Antalya, Türkiye, 15 - 18 Mayıs 2017 identifier identifier


Image denoising is the process of removing or reducing noise from an image. There have been many proposed methods and algorithms and each technique has its benefits and limitations. Mean filter is one of the simplest and easiest to implement smoothing filters. However, the main drawback of this method is its incapacity to preserve region boundaries and its information while smoothing an image. In this paper, the growing region segmentation based method is used to detect whether the pixels are part of the edges or the textures. In this way, we determine which neighbors should be included in the calculation for each pixel. The proposed method achieves the smoothing away of textures whilst retaining the edges. Experimental results show that the proposed method outperforms two well-known and widely-available image denoising algorithms in terms of PSNR (Peak signal-to-noise ratio) values.