Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Face image super-resolution reconstruction method based on morphological component analysis

A super-resolution reconstruction, morphology-based technique for image processing

Inactive Publication Date: 2013-02-06
SUN YAT SEN UNIV
View PDF2 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, the present invention can also be extended to simultaneously solve the problems of face image super-resolution and expression normalization

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
  • Face image super-resolution reconstruction method based on morphological component analysis
  • Face image super-resolution reconstruction method based on morphological component analysis
  • Face image super-resolution reconstruction method based on morphological component analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Such as figure 1 As shown, the face image super-resolution reconstruction method based on morphological component analysis includes the following steps:

[0058] (1) Upsampling the low-resolution input image by interpolation method to obtain the interpolated image;

[0059] (2) Using the morphological component analysis method to decompose the interpolated image obtained in step (1) into an overall high-resolution face image and an unsharp mask;

[0060] (3) Downsample the overall high-resolution face image obtained in step (2), and then subtract it from the low-resolution input image to obtain a low-resolution residual image, and then pass each of the low-resolution residual images Perform neighbor reconstruction on the image block at the face position to compensate the face detail information of the image to obtain a high-resolution residual image;

[0061] (4) Merge the high-resolution residual image in step (3) with the overall high-resolution face image obtained ...

Embodiment 2

[0080] The structure of this embodiment is the same as that of Embodiment 1 except for the following features: based on the multi-channel decomposition capability of the morphological component analysis method, the expression changes in the morphological component analysis are modeled by designing a dictionary, and the algorithm is extended to simultaneously process human face images Super-resolution and expression normalization. A face image super-resolution reconstruction method based on morphological component analysis, comprising the following steps:

[0081] (1) Set five basic facial expressions: eyes closed, frown, smile, surprise, mouth open. Through the method of principal component analysis, the five kinds of expressions are learned to obtain a dictionary of expressions that are not related to each other, as follows:

[0082] (1-1) For each expression, randomly select different face high-resolution images of this expression and the corresponding high-resolution image...

Embodiment 3

[0091] The structure of this embodiment is the same as that of Embodiment 1 except for the following features: In this embodiment, based on the face image super-resolution reconstruction method based on morphological component analysis, all low-resolution input images are processed by a classifier before processing Judge whether its human face has expression, if not, then adopt the method described in embodiment 1 to process, if have expression, then adopt the method described in embodiment 2 to process.

[0092] The present invention illustrates the effects of the present invention through the following experiments: the CES-PEAL-R1 face database is selected in the experiment, and the database contains 99,594 images of 1040 people. The position is simply calibrated. As shown in Table 1, 5 subsets of frontal images are used in two experiments, and some existing methods are also implemented for comparison:

[0093] Table 1. Contents of the 5 subsets of the CAS-PEAL-R1 database ...

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 face image super-resolution reconstruction method based on morphological component analysis. The method comprises the following steps: upsampling a low-resolution input image to acquire an interpolated image; acquiring a whole high-resolution face image from the interpolated image by a morphological component analysis method; downsampling the whole high-resolution face image and subtracting the whole high-resolution face image from the low-resolution input image to acquire a low-resolution residual image, and performing face detail information compensation by performing neighboring construction on an image block at each face position to acquire a high-resolution residual image; and combining the high-resolution residual image with the whole high-resolution face image to acquire the face image super-resolution result and finish resolution reconstruction. The invention also provides a method for performing face image super-resolution reconstruction and expression normalization simultaneously on the basis of morphological component analysis. By the method, detailed face detail information can be acquired, the ringing phenomenon is eliminated and high image quality is achieved.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method for super-resolution reconstruction of face images based on morphological component analysis. Background technique [0002] The task of image super-resolution is to infer a high-resolution image from one or more low-resolution input images. It is widely used in real life, especially the super-resolution reconstruction of face images has important applications in long-distance video surveillance or video processing. [0003] The traditional face image super-resolution method combines the classic prior knowledge unique to face images, and many face image super-resolution methods have emerged in the past ten years, such as constructing high-frequency components based on differential information, and based on principal components. Analysis (PCA) establishes an overall model to reconstruct high-resolution face images, and combines the overall model based on PCA and the local ...

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 Applications(China)
IPC IPC(8): G06T5/00G06T5/50G06K9/66
Inventor 赖剑煌梁炎
Owner SUN YAT SEN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com