Method and device for face image super-resolution reconstruction

A super-resolution reconstruction, face image technology, applied in image enhancement, image data processing, graphic image conversion and other directions, can solve the problems of low definition and poor effect of face images, recover detailed information and maintain global structure. , the effect of improving clarity

Active Publication Date: 2016-08-31
BEIJING INFORMATION SCI & TECH UNIV
View PDF1 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] At present, there is a face image super-resolution reconstruction technology that is based on the combination of the global parameter model and the non-parametric model of the local Markov random field to learn high-resolution face images. However, the reconstructed face images Less sharpness, poorer effect

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for face image super-resolution reconstruction
  • Method and device for face image super-resolution reconstruction
  • Method and device for face image super-resolution reconstruction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] see figure 1 , the present embodiment provides a method for face image super-resolution reconstruction, including:

[0046] 101: Divide both the test face image and the training face image into image blocks, wherein the resolution of the test face image is lower than the specified resolution, and the training face image includes a high-resolution training face image and a low-resolution image. A high-resolution training face image, the resolution of the high-resolution training face image is not lower than the specified resolution, and the low-resolution training face image has a resolution lower than the specified resolution;

[0047] 102: Divide all image blocks in the test face image into two categories according to smoothness, smooth blocks and non-smooth blocks, and then continue to divide each non-smooth block until there is no non-smooth block or non-smooth block after division. Stop dividing when the smooth block meets the preset condition;

[0048] 103: The t...

Embodiment 2

[0065] see figure 2 , the present embodiment provides a method for face image super-resolution reconstruction, including:

[0066] 201: Divide the test face image and the training face image into image blocks in an overlapping manner, and the number of image blocks in any training face image is the same as the number of image blocks in the test face image.

[0067] Wherein, the resolution of the test face image is lower than the specified resolution; the above-mentioned training face image includes a high-resolution training face image and a low-resolution training face image, and the training face image of the high-resolution The resolution is not lower than the above-mentioned specified resolution, and the resolution of the low-resolution training face image is lower than the above-mentioned specified resolution. The designated resolution can be set as required, which is not specifically limited in this embodiment.

[0068] For example, see image 3 , where Figure a is a...

Embodiment 3

[0150] see Figure 9 , the present embodiment provides a device for face image super-resolution reconstruction, including:

[0151] The division module 901 is used to divide both the test face image and the training face image into image blocks, wherein the resolution of the test face image is lower than the specified resolution, and the training face image includes high-resolution A training face image and a low-resolution training face image, the resolution of the high-resolution training face image is not lower than the specified resolution, and the low-resolution training face image has a low resolution at the specified resolution;

[0152] The adaptive module 902 is used to divide all image blocks in the test face image into two categories according to smoothness, smooth blocks and non-smooth blocks, and then continue to divide each non-smooth block until there is no non-smooth block after the division Or stop dividing when the divided non-smooth block meets the preset ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method and device for reconstructing a super-resolution facial image, and belongs to the field of image processing. The method comprises the step of dividing a tested facial image and a trained facial image into image blocks; the step of dividing the image blocks of the tested facial image into smooth blocks and non-smooth blocks; the step of continuing to divide each non-smooth block until there is no non-smooth block or the divided non-smooth blocks meet the preset conditions; the step of dividing the trained facial image into sub-blocks according to the same manner; the step of calculating reconstructed image blocks corresponding to all non-smooth sliding blocks in the tested facial image; the step of carrying out bicubic interpolation on all smooth blocks in the tested facial image to obtain corresponding reconstructed image blocks; the step of synthesizing the reconstructed images of the non-smooth blocks of the tested facial image and the reconstructed images of the smooth blocks of the tested facial image into a facial image to obtain the super-resolution reconstructed facial image of the tested facial image according to the position. The device comprises a dividing module, a self-adaptation module, a reconstructing module and a synthesizing module. According to the method and device, the definition of the reconstructed facial image is improved.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method and device for super-resolution reconstruction of a face image. Background technique [0002] Image super-resolution reconstruction (Super Resolution Reconstruction, SRR) is the process of using software to reconstruct one or more low-resolution images into a high-resolution image. Image super-resolution reconstruction has a wide range of applications, such as in video surveillance, medical images, remote sensing images and other fields. In real life, it is costly and technically difficult to improve the resolution through hardware technology. Therefore, it is of great significance to start from the software direction and post-process the collected images to improve the image resolution. Human face is a special type of image with high similarity. In recent years, in public security and other fields, video surveillance technology has become more and more popular, but bec...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/50G06T3/40G06K9/62
Inventor 曹林刘丹周汐
Owner BEIJING INFORMATION SCI & TECH UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap