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

Power line foreign body detection method based on light convolution neural network in low altitude aerial images

A convolutional neural network and foreign object detection technology, applied in the field of deep learning, can solve problems such as the limited effect of target recognition methods, achieve the effect of ensuring real-time performance and improving detection efficiency

Active Publication Date: 2018-10-12
BEIJING UNIV OF TECH
View PDF5 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Considering the increasing number of power line inspection images and the limited effect of traditional target recognition methods, this invention proposes a method for detecting foreign objects on power lines in aerial images based on multi-scale convolutional neural networks.

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
  • Power line foreign body detection method based on light convolution neural network in low altitude aerial images
  • Power line foreign body detection method based on light convolution neural network in low altitude aerial images
  • Power line foreign body detection method based on light convolution neural network in low altitude aerial images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] According to the above description, the following is a specific implementation process, but the protection scope of this patent is not limited to this implementation process.

[0022] Step 1: Convolutional neural network-based foreign object detection model construction on power lines

[0023] Step 1.1: Construction of Power Line Detection Model Based on Lightweight Convolutional Neural Network

[0024] The current deep learning object detection model has a wide range of application scenarios and can often detect thousands of types of objects, such as YOLO9000, which can detect 9418 categories. In the power line inspection scenario, the types of targets are extremely limited, mainly three types of power lines, balloons and kites. In this model, only power lines need to be identified, and its features are very limited. Currently, the existing deep learning target detection models are too specific for power line scenarios. Over-redundancy, while the lightweight model is ...

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

A power line foreign body detection method based on light convolutional neural network in low altitude aerial images belongs to the field of computer vision, and a real-time detection method for powerline foreign body in aerial images of unmanned aerial vehicles is studied. Firstly, a light power line detection model is constructed by using convolution neural network, and the depth characteristics of power lines in aerial images are calculated; then a multi-target power line foreign body detection model is constructed by using convolution neural network, and the prediction value of multi-scale targets is calculated by using depth features of convolution layers with different lengths and widths; finally, the power line detection model is used to filter the video frame without a power line,and the multi-target power line foreign body detection model is used to realize the real-time detection of power line foreign body in low-altitude aerial images on the video where a power line is detected.

Description

technical field [0001] Based on the deep learning technology, the present invention studies a real-time detection method for foreign objects on power lines in aerial images of drones. First, the convolutional neural network is used to construct a lightweight power line detection model, and the depth features of the power line in the aerial image are calculated; then, the convolutional neural network is used to construct a multi-target power line foreign object detection model, and convolutional layers of different lengths and widths are used to calculate multiple The predicted value of the scale target; finally, the power line detection model is used to filter the video frames without power lines. On the video with power lines detected, the multi-target power line foreign object detection model is used to realize real-time power line foreign object detection in low-altitude aerial images. The invention belongs to the field of computer vision, and specifically relates to techno...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V20/13G06N3/045Y04S10/50
Inventor 张菁王立元卓力梁西李昱钊
Owner BEIJING UNIV OF TECH
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