To search, Click below search items.


All Published Papers Search Service


Image De-noising using DMWT and Neural Network Thresholding


Omeed kamal khorsheed


Vol. 17  No. 2  pp. 67-73


Over the last few years, modern image processing technologies have spread all the aspects of our life. The observer can notice that information security, medical diagnosis, military communication, to name but a few, all depends on image processing techniques. However, a crucial part of this field is image filtering and denoising, being the first phase in image processing. On the other hand, there are many types of research proposed different techniques in the trial to return the noised image to its original status, which is commonly known as image denoising, most of them still lack the efficiency and accuracy of doing so. This proposed technique benefits from one of the highly efficient techniques, which is digital multi-wavelet thresholding technique, and enhanced its efficiency using feed forward neural network, which is used in order to reach the best thresholding values. After evaluating this novel technique by conducting several experiments, it proofed its efficiency and accuracy in defining the thresholding values accurately and restoring the noised image to its original status. Moreover, this technique proofed its efficiency and enhanced results over previous techniques..


Image De-noising, Digital Multi-wavelet Thresholding, Neural Network, Thresholding, Bayes Shrinkage.