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

Two-dimensional normalized Gaussian filter method for feature extraction of three-dimensional surface topography

A three-dimensional topography and topographic feature technology, applied in the field of image processing, can solve the problem of low accuracy of topographic feature recognition, and achieve the effect of improving the accuracy rate

Active Publication Date: 2019-11-15
HARBIN INST OF TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the existing extracted topography features have a wide distribution in amplitude, low-amplitude features are covered by high-amplitude features, resulting in low accuracy of topography feature recognition, and propose to use Two-dimensional normalized Gaussian filter method for feature extraction of three-dimensional surface topography

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
  • Two-dimensional normalized Gaussian filter method for feature extraction of three-dimensional surface topography
  • Two-dimensional normalized Gaussian filter method for feature extraction of three-dimensional surface topography
  • Two-dimensional normalized Gaussian filter method for feature extraction of three-dimensional surface topography

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0022] Specific implementation mode 1: The specific process of the two-dimensional normalized Gaussian filtering method used in the extraction of three-dimensional surface topography features in this embodiment is as follows:

[0023] Step 1. Set the input of the filter as the image three-dimensional shape f(u, v) and the cut length λ c , based on the resection length λ of the 3D topographic features of the image to be extracted c , calculate the Gaussian weight function g(x, y) of the two-dimensional Gaussian filter;

[0024] Step 2, establishing the template function bf(u,v) of the image three-dimensional shape f(u,v), avoiding the edge distortion problem of the filtering result;

[0025] Step 3. Make the Gaussian weight function g(x, y) of the two-dimensional Gaussian filter move point by point on the three-dimensional shape f(u, v) of the input image, and calculate the normalized filter when moving to (u, v) The result t(u, v), when all positions are moved, the obtained ...

specific Embodiment approach 2

[0026] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that the input of the filter in the step 1 is the image three-dimensional topography f(u, v) and the resection length λ c , based on the resection length λ of the 3D topographic features of the image to be extracted c , to calculate the Gaussian weight function g(x, y) of the two-dimensional Gaussian filter, the specific process is:

[0027] The Gaussian weight function g(x, y) of the two-dimensional Gaussian filter is expressed as:

[0028]

[0029] where α is the Gaussian filter constant, According to the Gaussian distribution, the value range of (x, y) is The number of matrix points after taking the value is recorded as N x with N y ; (x, y) is the point of the Gaussian weight function.

[0030] Other steps and parameters are the same as those in Embodiment 1.

specific Embodiment approach 3

[0031] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that in the step two, the template function bf(u, v) of the image three-dimensional topography f(u, v) is established to avoid the edge of the filtering result Distortion problem, the specific process is:

[0032]

[0033] The boundary is (u, v) is the point of the shape feature to be filtered.

[0034] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.

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 two-dimensional normalized Gaussian filtering method for three-dimensional surface topography feature extraction, and relates to a Gaussian filtering method for topography feature extraction. The invention aims to solve the problem that the existing extracted morphological characteristics have relatively wide distribution in amplitude and low-amplitude characteristics arecovered by high-amplitude characteristics, so that the recognition accuracy of the morphological characteristics is low. The method comprises the following steps of: 1, setting the input of a filteras a three-dimensional shape and a cutting length, and calculating a Gaussian weight function of the two-dimensional Gaussian filter based on the cutting length of a three-dimensional shape feature tobe extracted; 2, establishing a three-dimensional shape template function; And 3, enabling the Gaussian weight function of the two-dimensional Gaussian filter to move point by point on the input three-dimensional morphology, calculating a normalized filtering result when the Gaussian weight function moves to the point of the morphology feature to be filtered, and after the Gaussian weight function moves to a complete part of position, obtaining a matrix composed of the normalized filtering result as a filtering result. The method is applied to the field of image processing.

Description

technical field [0001] The invention is used in the field of image processing, and in particular relates to a Gaussian filter method for shape feature extraction. Background technique [0002] In industrial production, the surface of the machined parts implies topographic features of different scales. Through the extraction of these features, it can be used to analyze the processing problems in the production process of parts and the friction and wear problems in the use process. For example, different mechanical failure problems such as scratch wear and pitting corrosion will produce morphology features of different forms and scales. Analyzing these morphology features can help improve the production process and prolong the life of the product. Not limited to the industrial field, but in the field of criminal science, by extracting the three-dimensional topographical features of bullet traces and tool traces at the crime scene, it can help criminal investigators solve case...

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): G06K9/46
Inventor 佟明斯黄穗楚潘昀路赵学增
Owner HARBIN INST 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com