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

A Curve Noise Reduction Method Based on Active Contour Model

An active contour model and curve technology, applied in the field of curve noise reduction, to achieve the effect of simple algorithm, fast operation speed and noise elimination

Inactive Publication Date: 2020-03-31
SHANDONG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

is an example of its application in image processing, however no one in the prior art has used active contour models for noise removal

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 Curve Noise Reduction Method Based on Active Contour Model
  • A Curve Noise Reduction Method Based on Active Contour Model
  • A Curve Noise Reduction Method Based on Active Contour Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] A curve noise reduction method based on an active contour model, comprising the following steps:

[0035] 1) The iteration formula is defined according to the internal energy formula, and the experimental value of position x after the kth iteration is:

[0036] x k =[X k-1 +c 1 |ΔX k-1 | 2 +c 2 |ΔΔX k-1 | 2 ]; the total energy of the entire curve after the kth iteration is Among them, k is the number of iterations; x∈(n 1 , n 2 );

[0037] 2) The constraint condition is E out (k)-E 0 (k)≤T 1 ; where E 0 (k) is the total energy of the curve before iteration; T 1 It needs to be determined through experiments according to the specific situation. But for any problem, E out The output curves are as figure 2 As shown, so the relative difference formula of the total energy can be adopted, namely

[0038]

[0039] Therefore, δ Eout It can be set to 0.01 according to the precision requirement.

[0040] 3) When the iteration meets the constraints, stop th...

Embodiment 2

[0042] The curve noise reduction method based on the active contour model as described in Embodiment 1, the difference is that the constraints are x 0 is the experimental value at position x before iteration, T 2 is the second threshold set; in practical applications, T 1 The value is not easy to grasp, but for any problem, E out The output curves are as image 3 As shown, therefore, the relative difference formula of the total energy can be used, that is

[0043]

[0044] therefore, It can be set to 0.01 according to the precision requirement.

Embodiment 3

[0046] The curve denoising method based on the active contour model as described in embodiment 1, the difference is that c 1 = 0.3, c 2 = 0.5. c 1 ,c 2 The size of depends on the convergence speed, the larger the value of the coefficient, the faster the convergence.

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 involves a curve noise reduction method based on active contour model.The curve noise reduction method described in the present invention makes full use of the characteristics of the SNAKE model that can control the continuity and smoothness of the curve, and improve the algorithm for one -dimensional signal containing noise.Noise reduction.For the one -dimensional curve situation handled in this method, the nature of the internal energy control Snake itself, that is, control the power of continuity and smoothness contained in the curve contour containing the noise;Change is the judgment of whether iteration is terminated.

Description

technical field [0001] The invention relates to a curve noise reduction method based on an active contour model, and belongs to the technical field of curve noise reduction. Background technique [0002] Ideally, the physical quantity signal measured by the instrument does not contain noise. If the signal is a one-dimensional continuous signal, a one-dimensional curve corresponding to this one-dimensional signal can be drawn through measurement. However, due to the interference of the environment, the device error of the measuring instrument itself, and the interference caused by thermal noise, the actual measured curve is often superimposed with noise. The most common type of noise is white noise with an average value of 0, which is shown as Continuous burrs with ups and downs, that is, curve noise. Curve noise affects the judgment of the true value of the curve and the visual effect. [0003] Traditional curve denoising methods include filter-based methods or wavelet-bas...

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): G06K9/00
CPCG06F2218/04
Inventor 马昕
Owner SHANDONG UNIV