High-voltage bushing detection method based on Hu invariant moment and support vector machine (SVM)

A technology of support vector machine and detection method, applied in computer parts, character and pattern recognition, image data processing, etc., can solve the problems of complex background, interference of similar equipment, high-voltage bushing registration, etc., to achieve good detection performance, The effect of solving accuracy problems

Inactive Publication Date: 2018-04-20
JIANGSU ELECTRIC POWER RES INST +3
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since the background of the substation environment is more complex than other scenes (for example, it contains many objects such as high-voltage bushings, transformers, supports, and fences), it greatly increases the difficulty of feature extraction and pattern recognition for high-voltage bushings.
For example, due to the complex environment of the substation, the traditional image segmentation method cannot accurately segment the high-voltage bushing, and the interference between similar devices makes it difficult for the existing SIFT algorithm to accurately register the high-voltage bushing

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
  • High-voltage bushing detection method based on Hu invariant moment and support vector machine (SVM)
  • High-voltage bushing detection method based on Hu invariant moment and support vector machine (SVM)
  • High-voltage bushing detection method based on Hu invariant moment and support vector machine (SVM)

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] The high-voltage bushing detection method based on Hu invariant moment and support vector machine SVM disclosed in the present invention will be further analyzed below in conjunction with the accompanying drawings.

[0023] The high-voltage bushing detection method disclosed in the present invention can be roughly divided into a training phase and a detection phase. The purpose of the training phase is to obtain the best training model based on the captured images, and the purpose of the detection phase is to use the obtained best training model to detect the high-voltage bushing.

[0024] Such as figure 1 , the training phase of the high-voltage bushing detection method disclosed in the present invention roughly includes the following steps:

[0025] Step 1: image binarization threshold segmentation, which specifically includes the following steps:

[0026] Step 1.1: Perform grayscale conversion on the captured image to obtain a grayscale image. The conversion formu...

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 high-voltage bushing detection method based on Hu invariant moment and an SVM. The method comprises that a picture obtained by a tour inspection robot is preprocessed; afterthat the picture is converted into a grayscale map, a Niblack method is used to carry out binary-threshold segmentation on the picture, and interference of complex background is removed; according toshape characteristics of the bushing and an interference object, a morphological method is used to corrode and expand the picture, and the position of an area where the bushing exists possibly is obtained; after preprocessing, the Hu invariant moment is used to calculate the area where the bushing exists possibly in the picture; and a characteristic obtained by calculation is used to train the SVMand adjust parameters, and the SVM is used to identify and locate the SVM. Under the condition that the background of a transformer station is complex, the method is good in detection performance, and the problem that the high-voltage bushing may not be detected accurately in the complex environment is solved.

Description

technical field [0001] The invention belongs to a detection method of a key power equipment target of electric equipment, and in particular relates to a high-voltage bushing detection method based on Hu invariant moments and SVM (Support Vector Machine, ie Support Vector Machine). technical background [0002] In recent years, with the increase in the number of substations and the increase in the demand for power supply stability and the concept of smart power station construction, it is particularly important to detect substations through intelligent detection methods. As an important component of the transformer, the high-voltage bushing plays the role of support and insulation, and will withstand voltage during operation. Since the high-voltage bushing is exposed to the air and affected by rain and pollution, it is prone to cracks and damages, which will affect its insulation performance. Once the high-voltage bushing fails, it will seriously affect the normal operation ...

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): G06K9/38G06K9/46G06K9/62G06T7/00G06T7/136
CPCG06T7/0004G06T7/136G06T2207/20081G06T2207/10004G06V10/28G06V10/462G06V10/44G06F18/2411G06F18/214
Inventor 徐长福薄斌陶风波刘洋左国玉马蕾
Owner JIANGSU ELECTRIC POWER RES INST
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