Analysing Medical Image with Discrete Wavelet Transform Under Uncertainty

Medical image denoising is essential for improving the precision and efficiency of diagnostic treatments. The Discrete Wavelet Transform (DWT) and Neutrosophic Set (NS) theories are used in this paper to provide a comparative examination of denoising methods. The Median and Wiener Filter are two frequently used denoising techniques that are utilized to assess and compare the performance of the suggested methodology. Fuzzy Logic is a powerful tool that can help enhance image quality by considering the uncertainty in the data. The neutrosophic set is an extension of the fuzzy set. This paper proposes an image-denoising method based on the Haar wavelet transform implemented on a Neutrosophic Set (NS). It allows to handle the neutralities and indeterminacies in the data, enabling a more robust and accurate denoising process. The proposed method first denoises the image in the spatial domain using all three filter types, then translates it into a neutral Set and classifies it as True (T), indeterminate (I), or False (F). The Neutrosophic Set Entropy is used to assess the level of indeterminacy in the image. The suggested approach is compared with two other methods, the median filter and the wiener filter, using metrics such as RMSE and PSNR for Gaussian and Salt & Pepper noise with varying degrees of noise density. The Neutrosophic Set Entropy is also used to assess the level of indeterminacy in the image. The results show that the proposed method can reduce noise and produce high-quality images. Also, the performance analysis of filters in two different domains is analyzed.