Assembly multi-view change detection method and device based on feature matching

A technology of feature matching and change detection, which is applied in the field of image processing, can solve the problems of failure to realize the type identification of assembly errors and assembly errors, and achieve the effects of shortening the production cycle, high real-time performance, and reducing the failure rate

Active Publication Date: 2021-08-20
QINGDAO TECHNOLOGICAL UNIVERSITY
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The patent failed to realize the identification of assembly errors and types of assembly errors in the assembly process

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
  • Assembly multi-view change detection method and device based on feature matching
  • Assembly multi-view change detection method and device based on feature matching
  • Assembly multi-view change detection method and device based on feature matching

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] A feature-matching-based assembly multi-view change detection method, comprising the following steps:

[0059] S1. Acquire a first image and a second image, the first image is an image of the assembly at a previous moment, and the second image is an image at a later moment of the assembly. In this embodiment, the shooting angles of the assembly in the first image and the second image are different (that is, the angles between the camera and the assembly are different).

[0060] S2. Using the SIFT algorithm to perform feature point extraction and feature matching of the first image and the second image (SIFT algorithm, Scale Invariant Feature Transform, scale invariant feature transformation), to obtain the matching pair set and the first mismatch point of the first image set and the second mismatch point set of the second image, the mismatch point is a feature point that does not match another image feature point in the image, specifically:

[0061] S21. Extract featur...

Embodiment 2

[0075] Further, the step S3 is specifically:

[0076] Through the K-means clustering algorithm (K-Means algorithm) or other clustering algorithms, perform cluster analysis on the first mismatch point set, divide similar mismatch points into the same cluster, and obtain several first mismatch points clustering; cluster analysis is performed on the second set of unmatched points, and similar unmatched points are divided into the same cluster to obtain several second unmatched point clusters.

[0077] For example, K mismatch points in the first mismatch point set are randomly selected as centroids, and Euclidean distances from each mismatch point in the first mismatch point set to each centroid are calculated. The smaller the Euclidean distance, the higher the similarity between the mismatch point and the centroid. Sequentially divide each unmatched point into the cluster where the centroid with the smallest Euclidean distance is located, and update the centroid of the cluster w...

Embodiment 3

[0086] Assume that in step S3, a first region set to be matched including M first regions to be matched and a second region set to be matched including N second regions to be matched are obtained.

[0087] see Figure 10 , the step S4 is specifically:

[0088] S41. Select region k in the second region to be matched; perform feature matching on region k and n first regions to be matched respectively, and calculate matching degree P:

[0089]

[0090] Among them, T represents the number of matching pairs between the first region to be matched and the second region to be matched, A represents the number of feature points in the first region to be matched, and B represents the number of feature points in the second region to be matched. The larger the value of P, the greater the number of matching pairs between the first region to be matched and the second region to be matched, and the higher the similarity.

[0091] If the maximum value P among the several matching degrees ...

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 an assembly multi-view change detection method based on feature matching. The method comprises the following steps: S1, obtaining a first image and a second image; S2, performing feature point extraction and feature matching on the first image and the second image to obtain a matching pair set, a first mismatching point set of the first image and a second mismatching point set of the second image; S3, acquiring a first to-be-matched region set of the first image according to the first unmatched point set; acquiring a second to-be-matched region set of the second image according to the second mismatching point set; S4, performing feature matching on the first unmatched areas and the second unmatched areas one by one to obtain a plurality of matching results; and S5, outputting the change type of the assembly according to the plurality of matching results. According to the invention, the images of the assembly in different assembly processes and at different visual angles are obtained, and errors in the assembly process can be found in time by identifying the change areas in the images and judging the change type of the assembly, so that the reject ratio of products is reduced, and the production cycle of the products is shortened.

Description

technical field [0001] The invention relates to a feature-matching-based multi-angle change detection method and equipment for an assembly, belonging to the field of image processing. Background technique [0002] Assembly is an important production process of product manufacturing, which refers to the process of assembling and connecting mechanical parts according to design requirements. As product types continue to change, so does the difficulty of assembly. In the assembly process of complex mechanical products, once the errors in the assembly process (such as wrong assembly sequence, missed assembly, wrong assembly, etc.) are not detected in time, it will directly affect the assembly efficiency and service life of mechanical products. Therefore, it is necessary to detect the changing state of the mechanical assembly and find errors in the assembly process in time. [0003] Existing technologies, such as the patent "A deep learning-based assembly monitoring method, equi...

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): G06K9/62
CPCG06V10/757G06F18/23213Y02P90/30G06T7/001G06T2207/30164G06F18/23G06V10/443G06T7/0004
Inventor 陈成军岳耀帅李东年洪军
Owner QINGDAO TECHNOLOGICAL UNIVERSITY
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