Total variation image denoising method based on adaptive weighted edge detection
An adaptive weighting and edge detection technology, applied in the field of total variational image denoising, which can solve the problems of incomplete boundary information, inability to accurately measure the directionality of edge information, and edge blurring, so as to achieve more image details and eliminate boundary effects. , the effect of fast convergence
Pending Publication Date: 2021-09-07
DALIAN UNIV OF TECH
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] One of the problems in the research of image processing technology is the gradient information of the image. The gradient information can be used to extract the image boundary, but the pure gradient information cannot accurately measure the directionality of the edge information when describing the edge of the image, making the boundary information incomplete.
In addition, the edge strength is measured by using the structural metric of the image gradient information. However, the pure structural metric will be affected by noise during processing. In order to make the desired structural metric more robust, the common method is to Performs a smoothing filtering operation, but this operation results in blurring at the edges
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
[0063] The method of the present invention solves the very important image restoration problem in image processing, reads the image u containing noise in the computer, and obtains a value ranging from 0 to 255, and the data type is a value of uint8 type Matrix, for the convenience of iterative calculations, it is usually converted into double data.
[0064] STEP1: First, set the parameters and initialize the algorithm, and set the tolerance tol=10 -5 , the maximum iteration number iterMax=1000, k=1, and let u 0 =u, Parameter δ ∈ [0, 1];
[0065] STEP2: Next, according to the calculation formula of fuzzy edge complement and weighted structure tensor, take the degeneration parameters h and Both are 1, and the boundary metric is calculated
[0066] STEP3: Execute the iteration of the original variable u, the objective function of u is known from the following formula is differentiable, so the Gauss-Seidel iteration method can be directly used to solve it,
[0067]
[...
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 relates to a total variation image denoising method based on adaptive weighted edge detection, and relates to a total variation model for adaptive weighted edge detection. Local structure properties in an image are retrieved by using a structural metric-structure tensor capable of reflecting gradient information of the image in each direction, an edge detection operator with robustness about noise disturbance is obtained, the operator is substituted into a general total variation denoising model, distinguishing processing of an image in a smooth area and a boundary area is achieved, a self-adaptive total variation image denoising model about detected edge information is constructed, and during solving, a smooth item and a non-smooth item of a target function in the model are processed separately, and the function of the smooth item is measured by adopting a Bregman distance. The method is used for solving the problem of image restoration.
Description
technical field [0001] The invention belongs to the technical field of digital image processing, and relates to a full variation image denoising method based on adaptive weighted edge detection. Background technique [0002] With the increasing popularity of electronic computers and digital imaging equipment, more and more image processing appears and is applied in the fields of life, science and engineering. However, disturbances usually occur in the process of image acquisition and transmission. The observed image f is the perturbation and degradation of the real image u. Its model is as follows, [0003] f=Ku+ε, [0004] Among them, K represents a linear compact operator, and ε represents noise. [0005] One of the problems in the research of image processing technology is the gradient information of the image. The gradient information can be used to extract the image boundary, but the pure gradient information cannot accurately measure the directionality of the edge i...
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 IPC IPC(8): G06T5/00G06T7/13G06K9/46G06K9/62
CPCG06T7/13G06F18/22G06T5/77G06T5/70
Inventor 庞丽萍田玉铢王帅
Owner DALIAN UNIV OF TECH