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

Multi-target tracking method and tracking system for aerial images of unmanned aerial vehicle

A multi-target tracking and aerial image technology, which is applied in the field of image processing and computer vision, can solve problems such as false detection and similar target crosstalk, and achieve high-efficiency detection, high robustness and detection effect

Pending Publication Date: 2020-09-18
深延科技(北京)有限公司
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of algorithm mostly uses the traditional characteristics of the target. For the situation where the number of targets in the aerial video is large and the target is small, this type of algorithm often makes false detections during the re-detection process, resulting in crosstalk between similar 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
  • Multi-target tracking method and tracking system for aerial images of unmanned aerial vehicle
  • Multi-target tracking method and tracking system for aerial images of unmanned aerial vehicle
  • Multi-target tracking method and tracking system for aerial images of unmanned aerial vehicle

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] The present invention proposes a multi-target tracking method for unmanned aerial vehicle images, and the specific steps are as follows:

[0038] (1) Data enhancement (image translation, zooming, channel transformation, illumination change) is performed on the UAV aerial photography dataset.

[0039] (2) The image sequence I is input into the deep neural network Faster RCNN for target detection, and the detected result sequence I' is obtained.

[0040] (3) I' is sent to the multi-target tracking network IOU Tracker, and the IOU Tracker adds all the detection frames of the first frame in I' to the tracking queue Q t , and then take the second frame as the current frame F c , the F c All detection boxes in bbox_now and Q t All detection frames in bbox_pre do IOU calculations, and bbox_now gets the corresponding maximum IOU value IOU_max respectively.

[0041] (4) Set the threshold σIO U, if IOU_max>σIO U, then set F c The detection frame in is added to the tracking q...

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 multi-target tracking method and tracking system for aerial images of an unmanned aerial vehicle. The multi-target tracking method comprises the steps that a Faster RCNN detection network carries out target detection on an input image sequence; an IOU Tracker tracker performs multi-target tracking on the detected image sequence to form an initial tracking trajectory; theKCF tracker carries out updating compensation on the initial tracking trajectories, and forward updating and backward updating are carried out on each initial trajectory in sequence; the method comprises the following steps: designing a trace vote (trajectory vote) algorithm, and for a trajectory updated by a KCF, calculating the trace vote by using a greedy algorithm and an IOU (intersection-parallel ratio) to classify and fuse a plurality of trajectories. According to the invention, a high-robustness, high-precision and high-efficiency multi-target tracking effect can be realized under the condition that the number of targets in aerial images of the unmanned aerial vehicle is large and the target scale is small.

Description

technical field [0001] The invention relates to the fields of image processing and computer vision, in particular to a multi-target tracking method for aerial images of unmanned aerial vehicles. Background technique [0002] Multiple Object Tracking (MOT) aims to track objects of all classes of interest in a video sequence, and it plays a vital role in applications such as video surveillance and autonomous driving. In recent years, camera-equipped drones, or general-purpose drones, have been rapidly deployed in a wide range of applications, including agriculture, aerial photography, rapid delivery, and surveillance, among others. The images taken by UAVs have the characteristics of many types and quantities of targets, and most of them are small targets. Therefore, how to perform multi-object tracking on the visual data collected from these platforms is a challenging and worthy research problem. [0003] Multi-target tracking algorithms are divided into online tracking and...

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): G06T7/223G06K9/62
CPCG06T7/223G06T2207/10016G06T2207/20081G06T2207/20084G06T2207/30241G06F18/259G06F18/254
Inventor 陈海波
Owner 深延科技(北京)有限公司
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