Abstract
We present a new image restoration method by combining iterative VanCittert algorithm with noise reduction modeling. Our approach enables decoupling between deblurring and denoising during the restoration process, so allows any well-established noise reduction operator to be implemented in our model, independent of the VanCittert deblurring operation. Such an approach has led to an analytic expression for error estimation of the restored images in our method as well as simple parameter setting for real applications, both of which are hard to attain in many regularization-based methods. Numerical experiments show that our method can achieve good balance between structure recovery and noise reduction, and perform close to the level of the state of the art method and favorably compared to many other methods.
Highlights
Image restoration aims to compensate for or undo the defects that degrade an image
Another development based on block matching 3-D (BM3D) is sparse representation for image restoration, where the image is considered to be a combination of a few atomic functions taken from a certain dictionary and can be parameterized and approximated locally or non-locally by these functions [4]
These images are the subjects of a recent extensive investigation by an iterative decoupled deblurring BM3D algorithm (IDD-BM3D) [6], which is formulated based on the Nash equilibrium balance of two objective functions undertaking separate denoising and deblurring operations
Summary
Image restoration aims to compensate for or undo the defects that degrade an image. Degradation can come in many forms such as motion blur, noise, and camera defocus. BM3D algorithms are initially developed for collaborative filtering through a non-local modeling of images by collecting similar image patches in 3D arrays [2] They have recently been incorporated into image restoration for solving regularized inverse problems for image denoising as well as deblurring [3]. Another development based on BM3D is sparse representation for image restoration, where the image is considered to be a combination of a few atomic functions taken from a certain dictionary and can be parameterized and approximated locally or non-locally by these functions [4]. The formulation of IDD-BM3D image modeling in terms of the overcomplete sparse frame representation for image reconstruction has led to impressive restoration performance [6] This approach allows decoupling between deblurring and denoising by considering the optimization problem
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have