A Non-local Low-Rank Regularized Image Compressed Sensing Reconstruction Method
A technology of image compression and compressed sensing, which is applied in image coding, image data processing, instruments, etc., can solve the problem of consuming large resources and achieve good technical support and good reconstruction effect
Active Publication Date: 2018-12-07
苏州协同创新智能制造科技有限公司
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] With the continuous development of digital multimedia, the compression, storage and transmission of images, videos and other signals commonly used in daily life require a lot of resources
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0041] The present invention will be described in further detail below in conjunction with the accompanying drawings.
[0042] The technical idea of the present invention is as follows: firstly, the original image is preprocessed with cooperative rank reduction, which mainly includes three steps of non-local similarity block finding, low rank approximation and weighted average. Then the measured value is obtained through random measurement, and finally the original image is efficiently reconstructed from the measured value through the non-local low-rank regularized compressed sensing reconstruction algorithm (NLR-CS). Refer to attached figure 1 , the present invention uses non-local similarity block finding, low-rank approximation and weighted average as the main steps of cooperative rank-reducing preprocessing to preprocess the original image, and then perform compressed sensing measurement and non-local low-rank regularization reconstruction on the processed image , to ob...
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
Login to View More Abstract
The invention discloses a non-local low-rank regularization image compression perception reconstruction method based on cooperative rank reduction preprocessing. Firstly cooperative rank reduction preprocessing is performed on an original image, wherein cooperative rank reduction preprocessing mainly comprises three steps of finding-out similar blocks in a non-local manner, low-rank approximating and weighted meaning. Then a measured value is obtained through random measurement. Finally an original image is efficiently reconstructed from the measured value through a non-local low-rank regularization compression perception algorithm. The non-local low-rank regularization image compression perception reconstruction method is based on a matching relation between the reconstruction algorithm and prior information in compression perception and realizes matching between the prior information and the backend compression perception reconstruction algorithm. The invention provides the non-local low-rank regularization compression perception algorithm with higher efficiency based on cooperative rank reduction preprocessing, thereby obtaining better reconstruction effect in a current image compression perception reconstruction field, and supplying better technical support for practical techniques such as image compression, image storage and code transmission in practical application.
Description
technical field [0001] The invention belongs to the field of image compression sensing reconstruction, and relates to an image compression sensing reconstruction algorithm based on preprocessing and reconstruction algorithm matching, in particular to a non-local low-rank regularized image compression sensing reconstruction method based on cooperative rank reduction preprocessing. Background technique [0002] With the continuous development of digital multimedia, the compression, storage and transmission of images, videos and other signals commonly used in daily life consume a lot of resources. Digital image compression, storage and robust coding transmission based on compressed sensing all have broad practical application value. The reconstruction algorithm in compressed sensing is a core content, so it is of great significance to study an efficient compressed sensing reconstruction algorithm. [0003] Although the current compressed sensing reconstruction algorithm has de...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T9/00
CPCG06T9/00
Inventor 侯兴松镡云
Owner 苏州协同创新智能制造科技有限公司