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

An automatic classification system of aerial photographing line patrol images based on depth learning

A deep learning and image classification technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as high costs, reduce risks, save training costs, and improve data management efficiency.

Inactive Publication Date: 2019-01-11
FUZHOU UNIV
View PDF3 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The amount of UAV line inspection image data is huge, and manual data classification requires extremely high costs. How to automatically classify UAV line inspection image data according to component categories is a technical problem that needs to be solved at present

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
  • An automatic classification system of aerial photographing line patrol images based on depth learning
  • An automatic classification system of aerial photographing line patrol images based on depth learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The technical solution of the present invention will be specifically described below in conjunction with the accompanying drawings.

[0017] The present invention provides an automatic classification system for aerial line patrol images based on deep learning, comprising the following steps:

[0018] Step S1, establish the line inspection image classification image library and its label library: including UAV line inspection pictures of different types of components, all pictures are divided into eight categories: towers, foundations, insulators, grounding devices, auxiliary facilities, ground wires, Small fittings and large fittings; the auxiliary facilities include bird-proof facilities and tower identification plates, small fittings include fasteners such as bolts and nuts, and large fittings include anti-vibration hammers, wire clips, pressure equalizing rings and spacers; each image category is Including fault pictures and non-fault pictures, the number of the two ...

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 relates to an automatic classification system of aerial photographing line patrol images based on depth learning. The method comprises the following steps of: establishing a line patrolimage classification image library and a label library thereof; establishing a deep learning model, including VGGNet, ResNet, DenseNet, NasNet and other high performance classification models; training a classification model, and randomly performing data enhancement operation on the input data of each iteration in the training process, including rotation, filling cut and grayscale; integrating thefour trained models and obtaining the integrated model; setting classification rules, inputting the images to be detected into the integrated model, taking the three image categories with the first three confidence levels as the classification results of the images to be detected, and copying and storing the results to the servers of the three categories for archiving; continuously optimizing theclassification model, and periodically upgrading the performance of the automatic classification system.

Description

technical field [0001] The invention belongs to the fields of high-voltage transmission line inspection technology, image recognition technology, and machine learning technology, and specifically relates to an automatic classification system for aerial inspection line images based on deep learning. Background technique [0002] Transmission lines undertake the function of electric energy transmission in the power system, and the inspection and maintenance of transmission lines is the focus of the work of power departments in various places. In recent years, due to its economy, flexibility and safety, UAV inspection has gradually become one of the main means of inspection and maintenance of transmission lines. The images obtained by drone aerial photography can not only allow ground operators to directly confirm obvious line faults on site, but also can be brought back to the inspection and maintenance center and stored in the server for subsequent data diagnosis and analysis...

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): G06K9/62G06K9/66G06N3/04G06N3/08
CPCG06N3/084G06V30/194G06N3/045G06F18/241
Inventor 缪希仁刘欣宇江灏陈静
Owner FUZHOU 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com