Abstract
The new method for combining semisoft thresholding rules during wavelet-based data compression of images with multiplicative noise is suggested. The method chooses the best thresholding rule and the threshold value using the proposed criteria which provide the best nonlinear approximations and take into consideration errors of quantization. The results of computer modeling have shown that the suggested method provides relatively good image quality after restoration in the sense of some criteria such as PSNR, SSIM, etc.
Highlights
Images which are registered from different sources and which have to be transferred or archived can be distorted by specific noises having a multiplicative character
We can discuss about synthetic aperture radar (SAR) images with multiplicative noise known as speckle, infrared devices with fixed pattern noise (FPN) like unstable photo element’s voltage sensitivity, etc
We formed the library of test images containing 25 reference images which have been taken from the library described in [20]
Summary
Images which are registered from different sources and which have to be transferred or archived can be distorted by specific noises having a multiplicative character. Video sources and data links add their own noises during image formation and transferring. We can discuss about synthetic aperture radar (SAR) images with multiplicative noise known as speckle, infrared devices with fixed pattern noise (FPN) like unstable photo element’s voltage sensitivity, etc. There are a lot of methods and algorithms invented in last decades trying to effectively filter and/or compress noisy images. The tasks of filtering and compression are decided separately. Filtering and compression algorithms are not harmonized, that leads to new distortions and artifacts onto the images restored after compression. Being theoretically focused on an additive Gaussian noise model, any filtering algorithm inevitably leads to unsatisfactory results
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.