Unlock instant, AI-driven research and patent intelligence for your innovation.

Tower accessory defect inspection image checking method and system based on artificial intelligence

A technology of artificial intelligence and accessories, applied in image analysis, image enhancement, image data processing, etc., to achieve the effect of ensuring safe and stable operation

Pending Publication Date: 2020-12-25
PUYANG POWER SUPPLY COMPANY STATE GRID HENAN ELECTRIC POWER +1
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Aiming at the problem that traditional pattern recognition technology cannot identify defects and hidden dangers of small parts with high precision, the present invention proposes an artificial intelligence-based inspection system and method for inspection images of tower attachment defects

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
  • Tower accessory defect inspection image checking method and system based on artificial intelligence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: An artificial intelligence-based inspection method for inspection images of tower attachment defects, such as figure 1 shown, including the following steps:

[0027] S1, collect images of various types of accessories on the tower, and mark the peripheral frames of the accessories in the image, and divide the marked images into a training set and a test set;

[0028] Various types of accessories refer to spacers, anti-vibration hammers and pressure equalizing rings. The images of the accessories are captured by rotor drones, helicopters or manual shooting, covering various environments in the four seasons of spring, summer, autumn and winter. images; when marking the peripheral frame of the attachments in the image, the complete and unoccluded attachments in each picture in the training set are directly marked with their peripheral frames, and the incomplete or occluded attachments are not marked; while for In the test set, it is necessary to mark all atta...

Embodiment 2

[0041] Embodiment 2: An artificial intelligence-based system for inspecting images of defects in tower accessories, including an image acquisition module for inspections, which collects images of accessories, sends the images to the image recognition module, and the image recognition module performs image processing on the images The processing is sent to the feature extraction module; the image storage module stores a standard attachment image, and the feature extraction module performs attachment feature extraction on the processed image and sends it to the defect recognition model building module, and the defect recognition model building module is related to the image output module Connection, the defect recognition result determined by the defect recognition model building module is output through the image output module and displayed to the operation and maintenance personnel, which is convenient for viewing and timely maintenance.

[0042] The image output module is conn...

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 tower accessory defect inspection image checking method and system based on artificial intelligence, and the method comprises the following steps: collecting images of various types of accessories on a tower, marking peripheral frames for the accessories in the images, and dividing the marked images into a training set and a test set; establishing a convolutional neural network, inputting the training set into the convolutional neural network, and establishing an accessory defect identification model by using an R-CNN algorithm; setting a discrimination threshold, inputting the test set into the accessory defect identification model for defect identification, and comparing the marked images in the test set with a defect identification result to calculate the discrimination accuracy; and comparing the discrimination accuracy with a discrimination threshold, if the discrimination accuracy is greater than the discrimination threshold, collecting a new image to perform defect discrimination, otherwise, updating the image in the training set by adopting an elastic transformation algorithm, and retraining the model by utilizing the updated training set. Operation and maintenance personnel can be quickly and accurately helped to judge defect types, and defects can be conveniently and effectively eliminated in time.

Description

technical field [0001] The invention belongs to the technical field of power grid equipment state detection, and in particular relates to an artificial intelligence-based inspection system and method for inspection images of tower attachment defects. Background technique [0002] With the continuous innovation in the field of artificial intelligence technology, the use of new technologies such as image recognition to promote the development of power transmission inspection business has become a major issue in the power grid industry in recent years. The main content of intelligent identification of transmission line operation status includes three aspects: intelligent identification of transmission line operation status, intelligent identification of transmission line defects, and intelligent identification of transmission line hidden dangers. Traditional pattern recognition technology has been unable to meet the development needs of three-dimensional power transmission insp...

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): G06T7/00
CPCG06T7/0004G06T2207/20081G06T2207/20084
Inventor 邵震楚长鲲盛从兵谷延超王现吴述伟安金聚孙建军李新帅
Owner PUYANG POWER SUPPLY COMPANY STATE GRID HENAN ELECTRIC POWER