Image super resolution (SR) reconstruction method based on subspace projection and neighborhood embedding

Abstract

The invention discloses an image super resolution (SR) reconstruction method based on subspace projection and neighborhood embedding. The method is characterized by: using first and secondary subspace projection methods to project original high-dimensional data to a low-dimensional space, using dimension reduction feature vectors to show a feature of a low-resolution image block so that global structure information and local structure information of original data can be maintained; comparing a Euclidean distance between the dimension reduction feature vectors in the low-dimensional space, finding a neighborhood block which is most matched with the low-resolution image block to be reconstructed, using a similarity and a scale factor between the feature vectors to construct an accurate embedded weight coefficient so that a searching speed and matching precision can be increased; then constructing the similarity and the scale factor between the feature vectors, calculating the accurate weight coefficient and acquiring more high frequency information from a training database; finally, according to the weight coefficient and the neighborhood block, estimating the high-resolution image block with high precision, reconstructing the image which has the high similarity with a real object, which is good for later-stage real object identification processing.

Description

technical field [0001] The invention relates to an image processing method, in particular to an image super-resolution reconstruction method based on subspace projection and neighborhood embedding. Background technique [0002] Image super-resolution (SR) reconstruction technology refers to the use of one or more low-resolution (Low Resolution, LR) images captured at different observation angles, different observation times or different sensors in the same scene. Mutual information, digital image processing method is used to reconstruct a high resolution (High Resolution, HR) image. This reconstruction technology can estimate the lost high-frequency information through one or more low-resolution images under the conditions limited by hardware devices such as Charge Coupled Devices (CCD). In the fields of remote sensing satellites, military reconnaissance, medical imaging, security monitoring, and traffic management, image super-resolution reconstruction technology not only ...

AI Technical Summary

Problems solved by technology

Although the method of feature dimensionality reduction is used in Document 4, when the original low-resolution and high-resolution feature vectors are simultaneously projected onto the unified low-dimensional feature subspace, since the dimensionality of the original high-resolution feature vector is larger than that of the original low-resolution The large feature vector will inevitably damage the useful information in the original high-resolution feature vector, and will also have a bad effect on finding matching neighborhood blocks

In addition, the algorithms in Documents 1 to 4 all use local linear embedding algorithms when calculating the embedding weight coefficients, and the calculated embedding weight coefficients may be negative, which will affect the high-frequency information that the neighbor training high-resolution image blocks can provide. degree, and the computational complexity of the algorithm is high, and the operation speed is slow

Images