Unmanned aerial vehicle aerial image road extraction method based on deep learning

A technology of road extraction and deep learning, applied in neural learning methods, computer parts, instruments, etc., can solve the problems of inapplicability to complex scenes and weak robustness, and achieve robustness enhancement and detection speed improvement. Effect

Active Publication Date: 2018-08-17
西安因诺航空科技有限公司
View PDF2 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods only use simple artificial features, ar

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
  • Unmanned aerial vehicle aerial image road extraction method based on deep learning
  • Unmanned aerial vehicle aerial image road extraction method based on deep learning
  • Unmanned aerial vehicle aerial image road extraction method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0043] See figure 1 As shown, the present invention is a deep learning-based UAV aerial image road extraction method. First, the high-resolution UAV aerial image is divided into blocks, and then the divided image is quickly calculated by multi-threaded parallel computing. Graph model, then large-scale mean filtering is performed based on the integral graph model, and then the road segmentation model is obtained through the full convolutional neural network, and finally the road area is repaired to complete the road detection. The specific steps are as follows:

[0044] S1. Segment the high-resolution UAV aerial image;

[0045] For a drone aerial image with a resolution of 6000*4000, the image is divided into 1000*500 48 small blocks. For each image, the road segmentation result is obtained separately, and the image blocks are finally merged to obtain the final road segmentation result. When calculating the integral graph model of the divided image, it can ensure that the value ran...

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 an unmanned aerial vehicle aerial image road extraction method based on deep learning. Firstly the high-resolution unmanned aerial vehicle aerial image is partitioned; then anintegral graph model of the partitioned image is computed by using the multithread parallel computing method; then large scale mean filtering is performed based on the integral graph model; then a road segmentation model is obtained through the full convolutional neural network; and finally the road area is restored so as to complete road detection. The method has fast detection speed and has certain robustness for the road scale change and has higher detection accuracy than that of the conventional method.

Description

technical field [0001] The invention belongs to the technical field of image segmentation, and in particular relates to a method for extracting roads from UAV aerial images based on deep learning. Background technique [0002] In recent years, drone technology has developed rapidly. The UAV aerial photography system uses unmanned aerial vehicles as the carrier, equipped with modern digital cameras to quickly acquire low-altitude high-resolution images, and transmits the image data to the control platform in real time through the wireless network. The aerial photography system can photograph areas of interest, avoiding the blindness and ineffective work of conventional surveys, and has high mobility and flexible shooting time. It is widely used in territorial monitoring, disaster monitoring, resource exploration and other fields. One of the current main research directions of UAV is to use visual sensors to improve its autonomous navigation ability and scene understanding ab...

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): G06K9/00G06K9/34G06N3/04G06N3/08
CPCG06N3/08G06V20/182G06V10/267G06N3/045
Inventor 胡耀辉白霖抒成凯华韩姣姣马泳潮韦兴旺
Owner 西安因诺航空科技有限公司
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap