Image recognition detection method for train bottom bolt

A detection method and image recognition technology, applied in the field of image recognition, can solve the problems of poor compatibility and versatility, high false detection rate, uneven quality of bolt images, etc., and achieve good versatility and robustness. complex and varied effects

Active Publication Date: 2022-07-08
SOUTHWEST JIAOTONG UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Disadvantages of existing methods: This type of algorithm is generally poor in compatibility and versatility. In the background of this project, in the subway train bottom, the shooting environment is complex and changeable, and the quality of bolt images is also uneven. Using traditional Advanced image recognition technology will lead to high false detection rate and poor recognition effect

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
  • Image recognition detection method for train bottom bolt

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0029] An image recognition and detection method of a train underbody bolt of the present invention, the flow chart is as follows figure 1 shown, including the following steps:

[0030] Step 1: Use the labeimg tool to label and perform image enhancement operations, and use yolo-v4 for bolt target model training.

[0031] Step 2: Recognize the target of the bolt on the captured image, intercept the bolt, and use cv2.resize to unify the bolt image with a width of 800 and a height of 600, that is, the total pixel points are 480,000.

[0032] Step 3: Filter the bolt image, and use the cv2.bilateralFilter bilateral filter function in opencv2 for preprocessing. This filter can smooth the image and better protect the edges. After multiple comparison experiments, the parameters of the function are determined value, where the diam...

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 an image recognition detection method for train bottom bolts, and the method specifically comprises the steps: carrying out the image enhancement operation of a photographed image, carrying out the target recognition of a bolt, and carrying out the filtering processing of a bolt image; converting the image into an HSV space from an RGB space, then converting the image into a binary image, and carrying out image morphological operation, namely carrying out expansion and corrosion operation on the image; carrying out binary image edge detection, drawing a minimum circumscribed circle, drawing a minimum circumscribed rectangle on the basis of the circle, and extracting a bolt anti-loosening line by adopting a method of firstly establishing a color library and then screening thresholds in the library; calculating the relationship between every two circles by extracting parameters of circumcircles, and judging whether the group of thresholds is reasonable or not; unified comparison is carried out through a dictionary generation method, and then a result is output. According to the method, the image quality problem can be effectively solved, the problem that the shooting environment is complex and changeable can be effectively solved by using a self-created color screening algorithm, and the method has better universality and robustness.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to an image recognition and detection method of a train bottom bolt. Background technique [0002] In today's industrial field, bolts play an irreplaceable and important role in it, and bolt failure and loose detection are an essential and important work, but now industrial production and facilities are getting larger and more complex, and the number of bolts is also increasing. In the multiplication, relying on manual maintenance will not only consume a lot of manpower, but also make wrong judgments when faced with a large number of inspection points. The background of this project is the detection of the bottom of the subway train. There are thousands of bolt points, and the bolt shooting environment is complex and changeable. Therefore, a reliable bolt detection algorithm is urgently needed in the industrial field. [0003] Existing known technology: CN2015...

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): G06T7/00G06T7/62G06V10/28G06V10/44G06V10/772G06V10/764G06K9/62
CPCG06T7/0004G06T7/62G06T2207/20024G06T2207/20036G06T2207/30108G06F18/28G06F18/24Y02T10/40
Inventor 赵怡景杨然陆江关子凌李谨涵
Owner SOUTHWEST JIAOTONG UNIV
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