Face super-resolution reconstruction method based on feature transformation based on nearest feature line

Abstract

The invention discloses a nearest feature line based feature conversion face super-resolution reconstruction method, which is characterized in that face super-resolution reconstruction is performed by integrating a nearest feature line manifold learning based face super-resolution algorithm and a local feature conversion based face super-resolution algorithm, so that the method disclosed by the invention applies an idea of feature conversion compared with the nearest feature line manifold learning based face super-resolution algorithm, can also maintain most of the original information of an image while removing a lot of noise interference, and is mainly represented as better robustness to the noise; and the method improves the expression ability of a training sample library due to adoption of an idea of the nearest feature line compared with the local feature conversion based face super-resolution algorithm, enables a reconstructed target image to have better high-frequency local detail information and achieves an excellent image reconstruction effect.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a face super-resolution reconstruction method based on feature transformation based on nearest feature lines. Background technique [0002] Face super-resolution reconstruction is a method based on the observed low-resolution face image, using high-resolution image training library samples and low-resolution face image training library samples, to reconstruct the low-resolution face image to be reconstructed The most similar high-resolution face image; it can reproduce the local details of the face, achieve the purpose of enhancing the accuracy of face recognition, help improve the detection rate of public security organs, and protect the lives and property of the people. [0003] The face super-resolution algorithm based on local feature transformation introduces the idea of ​​partial face into the feature transformation super-resolution method, which improves the image...

AI Technical Summary

Problems solved by technology

[0003] The face super-resolution algorithm based on local feature transformation introduces the idea of ​​partial face into the feature transformation super-resolution method, which improves the image reconstruction effect to a certain extent, but due to the limitation of the size of the training sample library, the reconstruction effect Not ideal; Wuhan University Jiang Junjun and others proposed a face super-resolution algorithm based on the nearest feature line manifold learning, using the idea of ​​the nearest feature line to expand any two sample points of the same type into these two sample points Countless projection points on the characteristic line of the sample database solve the problem of sparse sample space distribution caused by too small a sample database, and greatly improve the expressive ability of the sample database. Rod requirements

Images