Scholarlink

In this chapter, we detail our image denoisingImage denoising technique proposed in literature. It is based on applying the self-organizing migration (SOMA)Self-Organizing Migration (SOMA) and an adaptive frequency median filter (AFMF)Adaptive Frequency Median Filter (AFMF). It consists firstly, in applying AFMF to the noisy image, \(I_{b}\) for obtaining a first version of the denoised image, \(I_{d1}\) which is considered as a clean image, \(I\) . The denoised image, \(I_{d1}\) , is employed as one input of an image denoising system based on SOMA. The latter is then applied for denoising the noisy image, \(I_{b}\) and a final version of the denoised image, \(I_{d2}\) , is obtained. This denoising system has two inputs which are \(I_{b}\) and \(I\) . The latter is not avalaible and we only have the noisy image, \(I_{b}\) , at our disposal. That’s why we have firstly applied the AFMFAdaptive Frequency Median Filter (AFMF) to \(I_{b}\) and a first version of the denoised image, \(I_{d1}\) , is obtained and is considered as the clean image, \(I\) . For ameliorating this proposed technique, this denoising system based on SOMASelf-Organizing Migration (SOMA) was replaced by the targeted image denoisingImage denoising (TID) system which is a more recent technique. The peak signal-to-noise ratio (peak-SNR) and structural similarity (SSIM)Structural Similarity (SSIM) were employed for the evaluation of the performance of the image denoising approaches introduced in this work. The results obtained from computing the PSNRPeak Signal-to-Noise Ratio (PSNR) and SSIM have proven the performance of those image denoisingImage denoising approaches. In fact, according to these results, these approaches outperform a number of image denoising techniques introduced in the literature.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Image Denoising Approaches Employing Adaptive Frequency Median Filter (AFMF)

  • 章节

  • 发表日期：

    2026年

查看全文
  • Mourad Talbi,
  • Brahim Nasraoui

摘要

In this chapter, we detail our image denoisingImage denoising technique proposed in literature. It is based on applying the self-organizing migration (SOMA)Self-Organizing Migration (SOMA) and an adaptive frequency median filter (AFMF)Adaptive Frequency Median Filter (AFMF). It consists firstly, in applying AFMF to the noisy image, \(I_{b}\) for obtaining a first version of the denoised image, \(I_{d1}\) which is considered as a clean image, \(I\) . The denoised image, \(I_{d1}\) , is employed as one input of an image denoising system based on SOMA. The latter is then applied for denoising the noisy image, \(I_{b}\) and a final version of the denoised image, \(I_{d2}\) , is obtained. This denoising system has two inputs which are \(I_{b}\) and \(I\) . The latter is not avalaible and we only have the noisy image, \(I_{b}\) , at our disposal. That’s why we have firstly applied the AFMFAdaptive Frequency Median Filter (AFMF) to \(I_{b}\) and a first version of the denoised image, \(I_{d1}\) , is obtained and is considered as the clean image, \(I\) . For ameliorating this proposed technique, this denoising system based on SOMASelf-Organizing Migration (SOMA) was replaced by the targeted image denoisingImage denoising (TID) system which is a more recent technique. The peak signal-to-noise ratio (peak-SNR) and structural similarity (SSIM)Structural Similarity (SSIM) were employed for the evaluation of the performance of the image denoising approaches introduced in this work. The results obtained from computing the PSNRPeak Signal-to-Noise Ratio (PSNR) and SSIM have proven the performance of those image denoisingImage denoising approaches. In fact, according to these results, these approaches outperform a number of image denoising techniques introduced in the literature.