Magnetic resonance imaging (MRI) techniques serve a important role in medical image processing. Various disturbances, namely Gaussian, salt and pepper, speckle noises, generally impact obtained images. To minimize noise in medical images for further analysis, various filtering algorithms are applied. Image translation from spatial (normal) domain to Neutrosophic Domain is a type of advanced image processing. This includes picture splitting into three images. Foreground object as a process of background subtraction, boundary object as a process of edge detection and background. Numerous digital filters, including the average, wiener, and median filters, are used in this work to remove various levels of noises that are present in MR images. The effect of the filtering techniques both in Normal Domain and Neutrosophic Domain are compared with the statistical parameter peak signal to noise ratio (PSNR) and root mean square error (RMSE).

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

Computational Analysis of Denoising Techniques in Spatial and Neutrosophic Domains

  • M. M. Shanmugapriya,
  • P. Divya,
  • S. Santhosh Kumar

摘要

Magnetic resonance imaging (MRI) techniques serve a important role in medical image processing. Various disturbances, namely Gaussian, salt and pepper, speckle noises, generally impact obtained images. To minimize noise in medical images for further analysis, various filtering algorithms are applied. Image translation from spatial (normal) domain to Neutrosophic Domain is a type of advanced image processing. This includes picture splitting into three images. Foreground object as a process of background subtraction, boundary object as a process of edge detection and background. Numerous digital filters, including the average, wiener, and median filters, are used in this work to remove various levels of noises that are present in MR images. The effect of the filtering techniques both in Normal Domain and Neutrosophic Domain are compared with the statistical parameter peak signal to noise ratio (PSNR) and root mean square error (RMSE).