A similar block accumulation image denoising method and system based on Bayesian tensor decomposition

A technology of tensor decomposition and similar blocks, which is applied in the field of similar block accumulation image denoising methods and systems, can solve difficult problems and achieve enhanced effects

Pending Publication Date: 2019-05-07
SHANGHAI JIAO TONG UNIV
View PDF6 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods all need to take specific statistical indicators of noise as input variables, which are generally difficult to obtain directly in actual images.
Many of these methods require complex parameter adjustment or need to use a large amount of data for training to better complete the denoising task in certain specific scenarios.

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
  • A similar block accumulation image denoising method and system based on Bayesian tensor decomposition
  • A similar block accumulation image denoising method and system based on Bayesian tensor decomposition
  • A similar block accumulation image denoising method and system based on Bayesian tensor decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] Such as figure 2 As shown, this embodiment includes several steps of preliminary denoising, accumulation, tensor decomposition, tensor reconstruction and restoration. :

[0022] Among them, tensor decomposition includes the following steps:

[0023] Step 1) Set the image block as a third-order tensor, that is, the three dimensions are spatial horizontal, vertical and color dimensions; then combine similar image blocks into a fourth-order tensor along the new dimension The specific steps include: the input image cube with noise can be decomposed into Among them: the elements in the tensor ε are It is 0-mean Gaussian white noise, and τ is the noise precision (precision), which is the reciprocal of the variance.

[0024] Step 2) The fourth-order tensor Factorized by tensor parallelism into:

[0025] Among them: o represents the outer product of the vector, is a one-dimensional vector, tensor matrix Is the Kruskal operator to represent the operation of the ...

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 similar block accumulation image denoising method and system based on Bayesian tensor decomposition. A plurality of image blocks are divided from an original image; similar image blocks are clustered to form a higher-order tensor; tensor decomposition is performed on the high-order tensor to obtain a joint probability of a tensor matrix and noise precision, a maximum log-likelihood model parameter and a hyper-parameter based on the joint probability are estimated approximately through an alternate iteration method and a denoised high-order tensor is obtained, and finally each image block in the denoised high-order tensor is restored to obtain a denoised low-order image.

Description

technical field [0001] The present invention relates to a technique in the field of image processing, in particular to a similar block accumulation image denoising method and system based on Bayesian tensor parallel factorization. Background technique [0002] The application of digital images in modern society has been quite popular, and people have higher and higher requirements for high-quality images. However, during the image acquisition process, due to the influence of the environment and the accuracy of the equipment, the acquired image data will contain various noises, which will affect the visual effect of the image. Reducing the influence of image noise and restoring the original information of the image as much as possible has a wide application prospect. Corresponding to various noises, traditional denoising methods include: Gaussian low-pass filtering, mean filtering, median filtering, wavelet transform denoising, etc. In recent years, methods based on image f...

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): G06T5/00
Inventor 招浩华张丽清赵启斌
Owner SHANGHAI JIAO TONG 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