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

Compressed sensing image reconstruction method combining bilateral total variation and non-local low-rank regularization

A bilateral full variation and image reconstruction technology, which is applied in image enhancement, image analysis, image data processing, etc., can solve the problems of edge smoothing and information degradation of restored images, and achieve the effect of enhancing texture details

Pending Publication Date: 2021-06-25
ZHEJIANG UNIV OF TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem of over-smoothing and information degradation in the edge of the restored image based on the non-local low-rank reconstruction algorithm, the present invention provides an improved method, that is, adding the bilateral full variation as the global information prior, and obtaining a A new compressive sensing image reconstruction method that combines bilateral full variation and non-local low-rank regularization can achieve more accurate and high-quality image restoration results

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
  • Compressed sensing image reconstruction method combining bilateral total variation and non-local low-rank regularization
  • Compressed sensing image reconstruction method combining bilateral total variation and non-local low-rank regularization
  • Compressed sensing image reconstruction method combining bilateral total variation and non-local low-rank regularization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0064] The present invention will be further described below in conjunction with the accompanying drawings.

[0065] refer to Figure 1 ~ Figure 4 , a compressive sensing image reconstruction method combining bilateral full variation and non-local low-rank regularization, including the following steps:

[0066] Step 1, input the original test image x in the computer, and all test images are grayscale images of 256×256 pixels;

[0067] Test image used x ref figure 2 shown.

[0068] Step 2, set the sampling rate rates, and arrange the test images into a one-dimensional vector form x∈R N×1 , generating a sampling matrix Φ∈R M×N , to generate a CS measure by randomly sampling the Fourier transform coefficients of the input image, resulting in a measure y ∈ R M×1 ;

[0069] y=Φx

[0070] Sampling Rate N=256×256=65536, take 10% or 0.1 sampling rate, M=6553. A sampling matrix Φ is generated in a computer, and Φ is a part of a Fourier matrix.

[0071] Step 3, take the obta...

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

A compressed sensing image reconstruction method combining bilateral total variation and non-local low-rank regularization comprises the following steps: 1, inputting an original test image x into a computer, wherein all the test images are gray level images with 256 * 256 pixels; 2, setting a sampling rate, arranging the test images in a one-dimensional vector form, generating a sampling matrix, and generating CS measurement by randomly sampling Fourier transform coefficients of the input images to obtain a measurement value; 3, inputting the obtained measurement value and the sampling matrix into a reconstruction model of a proposed joint algorithm as input, and iteratively solving a recovery image with 256 * 256 pixels by using an alternating direction multiplier method; and 4, selecting a CS reconstruction algorithm to perform subjective and objective comparison of the recovered image and the test image, and objectively evaluating algorithm reconstruction performance and selecting a peak signal-to-noise ratio and structural similarity as evaluation indexes. According to the invention, a more accurate and high-quality image restoration result can be realized.

Description

technical field [0001] The invention belongs to the technical field of image reconstruction algorithms, and relates to a compressed sensing image reconstruction method combining bilateral full variation and non-local low-rank regularization. Background technique [0002] In recent years, the theory of Compressed Sensing (CS) has been proposed, which is a new signal sampling method that breaks through the frequency limit of Nyquist sampling theorem. Compressed sensing theory points out that if the signal is sparse or sparsely representable in a certain transform domain, it can be achieved by sampling and data compression simultaneously by using a small number of random measurements generated by random Gaussian matrices or partial Fourier matrices. Perfect reconstruction of sparse signals. The compressed sensing method has the advantages of low sampling rate and high acquisition efficiency, and has been widely used in various fields including 3D imaging, video acquisition, en...

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): G06T11/00G06T7/00G06T7/13
CPCG06T11/008G06T7/0002G06T7/13G06T2207/10004
Inventor 张坤浩覃亚丽郑欢任宏亮胡映天
Owner ZHEJIANG UNIV OF TECH
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