Auto Parts Recognition Method Based on Spatial Shape Context Features

A space context and auto parts technology, applied in computer parts, character and pattern recognition, image analysis, etc., can solve problems such as matching errors, misleading auto repair workers, and inaccurate recognition, so as to improve operating efficiency and accuracy , the effect of saving time

Active Publication Date: 2019-03-29
DALIAN ROILAND SCI & TECH CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] With the continuous development of the automobile industry, the types of automobiles and the types of automobile parts have also increased. For auto repair workers, the human brain alone cannot accurately remember the model, price, scope of application, etc. of all automobile parts. information, there is an urgent need for a wearable device to help auto repair personnel identify auto parts
The target recognition algorithm of smart glasses is the most important. Correct recognition will bring unprecedented convenience to auto repair workers, and wrong recognition will mislead auto repair workers.
Existing auto parts recognition algorithms are generally based on two-dimensional image feature recognition. Two-dimensional image features are helpless for occlusion situations, because they cannot extract feature information of occlusion parts.
This leads to inaccurate recognition and wrong matching, which brings great trouble to the work of later auto repair workers.

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
  • Auto Parts Recognition Method Based on Spatial Shape Context Features
  • Auto Parts Recognition Method Based on Spatial Shape Context Features
  • Auto Parts Recognition Method Based on Spatial Shape Context Features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] The auto parts recognition method based on spatial shape context features is characterized in that it has the following steps: S1. constructing an offline auto parts feature library; S2. extracting the features of the auto parts to be identified online, and comparing them with the off-line auto parts feature library Feature matching and recognition of auto parts.

Embodiment 2

[0037] It has the same technical solution as in Embodiment 1, and more specifically: matching and recognition are based on spatial shape context features, and the spatial context features refer to extracting feature points of three-dimensional space car parts to be identified, and dividing the spatial regions to form The relative positional relationship of the feature points is used as the feature of the auto part.

Embodiment 3

[0039] Have the technical scheme identical with embodiment 1 or 2, more specifically: described step S1. The concrete method of constructing off-line auto parts feature library is:

[0040] S1.1 Take out an auto part from the auto parts library, and shoot it from multiple angles. The captured images have overlapping parts. After the shooting, the image data is sent to the cloud server for further processing;

[0041] S1.2. Extract the SIFT feature points of each car part image collected, match the feature points with the next image, and use RANSAC to remove the mismatch points;

[0042] S1.3. Reconstruct the feature points extracted according to S1.2 according to the three-dimensional reconstruction principle of sequence images to obtain a three-dimensional point cloud, and map the n two-dimensional feature points extracted in S1.2 to a three-dimensional space to form a three-dimensional space Feature point set P{p 1 ,p 2 …p n}.

[0043] S1.4. Through the spatial shape con...

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

An auto parts recognition method based on spatial shape context features, which belongs to the field of auto parts recognition, is used to solve the problem of auto parts recognition. The features are extracted online, and are matched and identified with the auto parts in the offline auto parts feature library. The effect is that the incomplete extraction of image shape features caused by single-direction image acquisition is avoided.

Description

technical field [0001] The invention belongs to the field of recognition of auto parts, in particular to an auto part recognition method based on spatial shape context features. Background technique [0002] With the continuous development of the automobile industry, the types of automobiles and the types of automobile parts have also increased. For auto repair workers, the human brain alone cannot accurately remember the model, price, scope of application, etc. of all automobile parts. information, there is an urgent need for a wearable device to help auto repair personnel identify auto parts. The target recognition algorithm of smart glasses is the most important thing. Correct recognition will bring unprecedented convenience to auto repair workers, and wrong recognition will mislead auto repair workers. Existing auto parts recognition algorithms are generally based on feature recognition of two-dimensional images. Two-dimensional image features are useless for occlusion ...

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): G06T7/00G06K9/46G06K9/62
CPCG06T2207/30108G06T2207/10004G06V10/462G06V10/757
Inventor 田雨农王哲周秀田于维双陆振波
Owner DALIAN ROILAND SCI & TECH CO LTD
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