Image restoration and denoising method and system

Abstract

The invention discloses an image repairing and denoising method and system. For a designated original training image, the method includes the steps of decomposing a designated training sample image data matrix into a combined low-rank and sparse principal component feature coding matrix and sparse error matrix through the convex optimization technology by introducing the combined low-rank and sparse matrix decomposition concept, and determining a combined low-rank and sparse main component feature and error matrix of training image sample data according to the low-rank and sparse features of the training image sample data so that repairing and denoising can be conducted on original images probably containing errors and images subjected to repairing and denoising can be obtained. By means of the method and system, robust low-rank and sparse features of data are sufficiently considered while feature depiction is conducted on image data, defects in the prior are overcome, and image repairing and denoising performance and model robustness are improved.

Description

technical field

[0001] The invention relates to the technical fields of computer vision and image processing, in particular to an image restoration and denoising method and system. Background technique

[0002] In a large number of practical applications, real data can be described by high-dimensional attributes or features, such as visual images, but high-dimensional data often contains a lot of redundant information or noise. Therefore, how to perform effective image restoration and how to effectively describe images through feature learning or low-rank, sparse coding techniques has attracted extensive attention in recent years.

[0003] Feature extraction aims to obtain descriptive and compact features through mapping or transformation methods, and realize the transformation from high-dimensional to low-dimensional. PCA (Principal Component Analysis) is one of the most representative unsupervised feature learning models. The specific operation is: for a given data matri...

Images