Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Small-scale equipment part detection method based on weak supervision collaborative learning in open scene of electric power field

A technology of equipment components and detection methods, applied in the field of smart grid, can solve the problems of decreased detection speed, slow detection speed, low efficiency, etc., to achieve the effect of improving speed and accuracy, and enhancing learning ability

Active Publication Date: 2020-07-24
SHANDONG UNIV +3
View PDF9 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the inspection of power transmission equipment in my country mainly adopts traditional manual inspection and automatic inspection by drones, which have the following problems: heavy workload, low efficiency, and large lag in fault judgment
However, these two types of network models have their own advantages and disadvantages. The detection accuracy of the neural network model based on the candidate area is high, but the detection speed is slow; The number of grids has a great relationship. If you want to detect small targets, you often need to divide more grids, but it will cause the detection speed to drop rapidly.
Although the detection accuracy of the neural network model based on the candidate area is high, it often does not have a high accuracy guarantee for the detection of small targets.

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
  • Small-scale equipment part detection method based on weak supervision collaborative learning in open scene of electric power field
  • Small-scale equipment part detection method based on weak supervision collaborative learning in open scene of electric power field
  • Small-scale equipment part detection method based on weak supervision collaborative learning in open scene of electric power field

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0070] A small-scale equipment component detection method based on weakly supervised collaborative learning in an open scene in the electric power field, including the following steps:

[0071] S1: Preprocessing the images in the power open scene: use the annotation tool to annotate the normalized graphics;

[0072] S2: Extract image information and feature fusion: extract feature maps containing different scales of pictures, use ResNet's conv1-conv4 convolutional layer for feature extraction, and construct a feature pyramid between conv3 and conv4 convolutional layers after obtaining features; The purpose of inventing and constructing a feature pyramid is to enrich the extracted feature information, and at the same time increase the feature extraction time. Therefore, the research experiment found that when only the pyramid is constructed between the conv3 and conv4 convolutional layers, the richness of feature information and the extraction speed can be achieved. Trade-off; ...

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 small-scale equipment part detection method based on weak supervision collaborative learning in an electric power field open scene, and the method comprises the steps: carrying out the fusion of shallow features and deep features through employing a feature pyramid based on the features of a small target of an equipment part, and obtaining richer information. After the extracted multi-scale features are input into the candidate region generation network, candidate regions under different scale features can be generated, and the processing range of the strong and weaksupervised learning network is divided according to the scale sizes of the candidate regions, so that the high performance of the strong supervised sub-network and the collaboration of the weak supervised sub-network are brought into full play. The time cost is reduced to a great extent, and the efficiency and the precision are balanced. Meanwhile, a detection framework different from a classic Faster R-CNN model is used for detecting the target, and the precision and speed of small target detection are improved at the same time.

Description

technical field [0001] The invention discloses a small-scale equipment component detection method based on weak supervision collaborative learning in an open scene in the electric power field, and belongs to the technical field of smart grids. Background technique [0002] Electricity is one of the indispensable sources of energy in the production and life of human society. With the large-scale growth of transmission lines and the increasing number of transmission equipment, the safety inspection of equipment, especially the timely monitoring of equipment component defects, is becoming more and more important. At present, the inspection of power transmission equipment in my country mainly adopts traditional manual inspection and automatic inspection by drones, which have the following problems: heavy workload, low efficiency, and large lag in fault judgment. [0003] For this reason, research and development have been started in this technical field: using neural network lea...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08G06Q50/06
CPCG06Q50/06G06N3/08G06V2201/07G06N3/045G06F18/253G06F18/24
Inventor 聂礼强郑晓云战新刚姚一杨陈柏成尹建华
Owner SHANDONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products