Unlock instant, AI-driven research and patent intelligence for your innovation.

Compressed sensing image decoding method based on non-uniform quantizing noise model

A technology of quantization noise and compressed sensing, applied in the field of image decoding, which can solve the problems of low quality of reconstructed images at the decoding end and suboptimal signal reconstruction results.

Active Publication Date: 2013-06-05
BEIJING UNIV OF TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is an obvious defect in the JPEG-based image codec: when part of the frequency domain information at the encoding end is lost, the quality of the reconstructed image at the decoding end will be significantly reduced
The above application applies compressive sensing theory to image decoding, but it is based on a uniform quantization noise model, that is, the error limit ε is fixed for the image signal x, and the result of such signal reconstruction is not optimal

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 decoding method based on non-uniform quantizing noise model
  • Compressed sensing image decoding method based on non-uniform quantizing noise model
  • Compressed sensing image decoding method based on non-uniform quantizing noise model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The compressed sensing decoding method is divided into two important steps of observation and reconstruction. The image compression decoding standard consists of two links: encoding and decoding. The present invention regards the transformation in encoding as CS observation, and the transformation matrix is ​​equivalent to observation matrix, and replaces the inverse transformation in decoding with CS reconstruction, and ε is an error caused by quantization noise. Obviously, if the error ε can be accurately reconstructed, the convex set optimization method will obtain better reconstruction results than the inverse transformation.

[0053] It should be noted that in the specific implementation, the gradient sparse model is used to sparse the image, that is, the image signal x adopts the gradient sparse basis to make TV(x) sparse. In the following description, TV(x) means that x is in the gradient Sparse in meaning. Of course, any sparse base that can make the image spa...

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 relates to a compressed sensing image decoding method based on a non-uniform quantizing noise model, comprising the following steps of: (1) establishing the non-uniform quantizing noise model; (2) establishing a mapping relation between a quantizing step q and a margin of error Epsilon; (3) estimating the non-uniform quantizing margin of error (Epsiloni) corresponding to each pixel according to a quantization matrix Q and the step (2), wherein i =1,......,N; and (4) carrying out the reconfiguration of the compressed sensing on an image signal x. Because a compressed sensing theory is applied to the image decoding based on the non-uniform quantizing noise model, the invention can better compensate noise in the decoding process, thereby greatly improving the reestablishing quality of a CS (Counter Strike).

Description

technical field [0001] The invention relates to an image decoding method, in particular to an image decoding method based on compressed sensing. Background technique [0002] The image codec standard JPEG has been widely used. This type of codec eliminates the redundant components of the signal through the converter and quantizer at the encoding end, and reconstructs the signal through inverse transformation and inverse quantization at the decoding end. However, there is an obvious defect in the JPEG-based image codec: when part of the frequency domain information is lost at the encoding end, the quality of the reconstructed image at the decoding end will be significantly reduced. It is even more important to note that the quantizer and dequantizer of JPEG can introduce serious quantization noise, and this standard noise reduction function is very limited. [0003] The two quantization tables of JPEG act on 8×8 blocks. Its quantization step size takes into account the sensi...

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): H04N7/26H04N7/30H04N19/124H04N19/44
Inventor 施云惠尹宝才王开帅张臻
Owner BEIJING UNIV OF TECH