Hybrid image denoising in multiresolution wavelet domain

Recently, with the explosion in the number of digital images captured every day in all life aspects, there is a growing demand for more detailed and visually attractive images. However, the images taken by current sensors are inevitably degraded due to the noise caused by the acquisition, transferring, compression, encoding, and storage processes, including retrieval process into the various fields, such as medical, astrophysics, weather forecasting, etc., which contributes to impaired visual image quality. Therefore, more work is needed to reduce noise by preserving the textural, information, and structural features of the digital image. So far, there are different techniques for reducing noise that various researchers have done. Each technique has its own advantages and disadvantages. In this work, and to denoise the blurred image caused by Additive White Gaussian Noise (AWGN) in the digital images field, a new approach is designed and implemented as a noise removal system aimed to study the application of spatial domain filters with working mechanisms based on the multiresolution wavelet domain threshold value. Various types of wavelet transform and threshold techniques have been tested to achieve improved results of the image noise reduction process by distinguishing and removing the noise from the affected pixel units, and the wavelet decomposition is applied with a three-levels of analysis. Hard and soft threshold techniques and spatial filters were applied for each of the high-frequency and low-frequency sub-images. Then, an inverse wavelet transform was applied to the denoised image. Finally, the performance metrics Peak Signal to Noise Ratio (PSNR) is calculated to estimate the evaluation of the suggested method. Experimental evaluation outcomes of the proposed method reveal a better improvement of image quality concerning minimizing noise and edge preservation than the results of the related works instead of using a threshold-based WT domain or spatial domain filters separately. Studying the application of a mechanism of image denoising as a hybrid system to applying it to the noisy image with a mixed type of noise can be considered future work.