Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Contact network insulator detection method based on reconstruction and classification convolution self-coding network

An insulator detection and convolutional self-encoding technology, applied in neural learning methods, biological neural network models, analysis materials, etc., can solve problems such as background interference, complex images, and complex image processing techniques, and improve robustness and robustness. The effect of stability, removal of background interference, and good decision-making reference

Active Publication Date: 2020-06-19
SOUTHWEST JIAOTONG UNIV
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above methods have their own disadvantages. For example, GAN can only judge whether there is a fault, but not the distribution of the fault, and is easily affected by background interference. The AE network has functional redundancy, etc.
Since the images of catenary support and suspension devices collected on site are generally complex, and the image processing technology adopted is relatively complex, there is an urgent need for a simple and rapid image detection algorithm to quickly locate and detect the fault status of insulators

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
  • Contact network insulator detection method based on reconstruction and classification convolution self-coding network
  • Contact network insulator detection method based on reconstruction and classification convolution self-coding network
  • Contact network insulator detection method based on reconstruction and classification convolution self-coding network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0124] collected by Figure 13 Take the image as an example, the details are as follows:

[0125] 1. Positioning segmentation. Such as Figure 14 As shown, using the Mask-RCNN convolutional neural network can accurately locate the insulator area and segment it from the background.

[0126] 2. Rotate the image. Such as Figure 15 As shown, the convex hull is searched by the Graham-Scan algorithm, and the minimum circumscribed rectangle is calculated for the insulator mask area to remove the background interference to the maximum extent. The inclination angle of the circumscribed moment is calculated according to the rectangular coordinates, the insulator is adjusted to be horizontal by coordinate transformation, and the image of the horizontal insulator is obtained.

[0127] 3. Outlier detection. Such as Figure 16 As shown, in order to avoid the interference of noise, outlier detection based on K-Nearest Neighbor (KNN, kNearest Neighbor) is used to remove noise.

[012...

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 contact network insulator detection method based on reconstruction and a classification convolution self-encoding network. The method specifically comprises the following steps: 1, imaging a high-speed railway contact network supporting and suspending device; 2, establishing a sample data set of the insulator, and carrying out insulator target detection and segmentation;3, adjusting the insulator to be horizontal by utilizing coordinate transformation, removing noise by utilizing outlier detection, carrying out edge detection on the insulator, then carrying out quadratic function fitting, cutting the acquired insulator images one by one, and finally establishing an insulator piece data set; 4, building a reconstruction and classification convolution automatic coding network, judging whether there is an insulator misclassification or not, and extracting an insulator fault region; 5, clustering the separated foreground images, and establishing a fault judgmentcriterion according to a clustering result; whether the insulator fails or not is judged by setting a threshold value, and the fault level is further evaluated. The detection result is objective, realand accurate, and the defects of a traditional detection method are overcome.

Description

technical field [0001] The invention belongs to the technical field of high-speed railway image intelligent detection, and in particular relates to a catenary insulator detection method based on a reconstruction and classification convolutional self-encoding network. Background technique [0002] The catenary is one of the key components in the high-speed railway system, which plays a vital role in the stability and safe operation of the railway. However, due to the complex and harsh operating environment, the components of the catenary are easily damaged, which may cause interruption of train operation and endanger the safety of passengers. In the electrified railway power supply system, the arm support device mainly includes inclined arm, horizontal arm (tie rod), rod insulator and related parts. The rod insulator is used to suspend and support the inclined arm and the horizontal arm and keep the contact wire electrically insulated from the grounding body. The oblique ar...

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): G06K9/32G01N21/88G01N21/95G06K9/34G06K9/62G06N3/04G06N3/08
CPCG01N21/95G01N21/8851G06N3/08G06V10/267G06V10/25G06N3/045G06F18/24147G06F18/2415
Inventor 刘志刚刘文强王惠李昱阳杨成
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
Eureka Blog
Learn More
PatSnap group products