A recommender system to restore images with impulse noise
This is an incremental improvement for image processing applications, focusing on a specific noise type with identified noisy pixels.
The authors tackled image restoration for impulse noise by developing a collaborative filtering recommender system based on a new color image representation, achieving image quality improvements as measured by statistics on a standard database.
We build a collaborative filtering recommender system to restore images with impulse noise for which the noisy pixels have been previously identified. We define this recommender system in terms of a new color image representation using three matrices that depend on the noise-free pixels of the image to restore, and two parameters: $k$, the number of features; and $λ$, the regularization factor. We perform experiments on a well known image database to test our algorithm and we provide image quality statistics for the results obtained. We discuss the roles of bias and variance in the performance of our algorithm as determined by the values of $k$ and $λ$, and provide guidance on how to choose the values of these parameters. Finally, we discuss the possibility of using our collaborative filtering recommender system to perform image inpainting and super-resolution.