Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for de-noising of balanced morphology filter image optimized by particle swarm

A particle swarm optimization and morphological filtering technology, applied in the field of image processing, which can solve the problems of image information loss, inability to overcome morphological deficiencies, and inability to achieve results.

Active Publication Date: 2014-03-19
JIANGSU UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Although the existing morphological filtering methods can improve the image processing effect to varying degrees, they cannot overcome the inherent shortcomings of morphology.
Due to the inherent operational characteristics of morphological core operations erosion and dilation, that is, morphological erosion and dilation are both extreme operations. This extreme operation is likely to cause image information loss while removing noise, especially for images with low signal-to-noise ratio. Denoising, can not achieve satisfactory results

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
  • Method for de-noising of balanced morphology filter image optimized by particle swarm
  • Method for de-noising of balanced morphology filter image optimized by particle swarm
  • Method for de-noising of balanced morphology filter image optimized by particle swarm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] A particle swarm optimized balanced morphological filter image denoising method, comprising:

[0092] ①Set the input image as f, and its pixel size is W*H;

[0093]Define the zero square matrix unit structure element SE, whose size is n*n;

[0094] The equilibrium erosion operation is defined using the unit structure element SE: , that is, the collection The median value of the inner gray value is used as the gray value of the input image pixel point (i, j); wherein, the value range of i is [0, W-n], the value range of j is [0, H-n], and h takes The value range is [0,n-1], and the value range of k is [0,n-1];

[0095] The equilibrium expansion operation is defined using the unit structure element SE:

[0096] , that is, the collection The median value of the inner gray value is used as the gray value of the input image pixel point (i, j); wherein, the value range of i is [0, W+n-2], and the value range of j is [0, H +n-2], the value range of h is [0,n-1], and...

Embodiment 2

[0103] On the basis of Example 1, in the step 5, the peak signal-to-noise ratio PSNR is used as a cost function, and the particle swarm optimization technique is used to update the particle velocity V and the particle position X to obtain the globally optimal particle The method of location includes the following steps:

[0104] a: Define the historical optimal position of the i-th particle as P i =[P i1 ,P iD ], the peak signal-to-noise ratio PSNR of the balanced expansion image is the individual extremum P id , the highest peak signal-to-noise ratio PSNR among the m particles is the global limit g id , limiting the maximum flying speed of particles to be V max , the current number of iterations is t, and the initial value of t is 0;

[0105] b: According to the position and velocity of the current particle, according to the formula (1)

[0106] update the velocity of the particle;

[0107] Press (2)

[0108] Update ...

Embodiment 3

[0115] The image is processed on the basis of Embodiment 1 and Embodiment 2.

[0116] Such as figure 1 , the input image used by the present invention for image noise removal, the image is the standard test image "lena" plus impulse noise with a noise density of 15%, and the image size is 256*256 pixels.

[0117] Such as figure 2 , when the present invention is implemented, the number of particle swarms is set m=10, D=10, v max =1, the obtained output image has a peak signal-to-noise ratio PSNR=29.432db, and the optimal unit structural element SE has n=3.

[0118] Such as image 3 , the input image used by the present invention for image noise removal, the image is a standard test image "lena" plus impulse noise with a noise density of 40%, and the image size is 256*256 pixels.

[0119] Such as Figure 4 , when the present invention is implemented, the number of particle swarms is set m=10, D=10, v max =1, the obtained output image has a peak signal-to-noise ratio PSNR=...

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 method for de-noising of a balanced morphology filter image optimized by a particle swarm, comprising the following steps of: firstly, defining an unit structure element of a zero square matrix, a balanced erosion operation and a balanced expansion operation; secondly, updating speeds and positions of particles by a particle swarm optimization technology after a peak signal-to-noise ratio is taken as a cost function, so as to enable a transformed value of the global optimum particle position to be as the size of the unit structure element; and finally, performing the balanced erosion operation on an image by the optimized unit structure element, then performing the balanced expansion operation, and outputting the image after the balanced expansion operation. The method for de-noising of the balanced morphology filter image optimized by the particle swarm, disclosed by the invention, overcomes the inherent shortages in an extreme value operation used by the present morphology, can adaptively obtain the size of the unit structure element according to the degree of pollution on the image caused by noise and effectively remove pulse noise in the image, as well as can be widely used in the fields of image processing, licence plate information extracting, edge detecting etc.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a particle swarm optimized balanced shape filter image denoising method. Background technique [0002] In the process of image generation and transmission, various noises are often introduced. These noises not only destroy the real information of the image, but also seriously affect the visual effect of the image. Images disturbed by noise can be filtered out by linear or nonlinear filtering methods. Since image details are reflected as high-frequency components in the frequency domain, they are easily confused with high-frequency noise. Therefore, how to maintain image details and effectively filter noise has always been a key issue in image processing. Morphological filtering belongs to nonlinear filtering, which is a representative and promising filter at present, and its theoretical basis is mathematical morphology. [0003] Although the existing morpho...

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): G06T5/00
Inventor 朱幼莲黄成程钦倪福银刘舒祺许致火
Owner JIANGSU UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products