Supercharge Your Innovation With Domain-Expert AI Agents!

A Fast Automatic Target Detection Method

A target detection and target technology, applied in image analysis, image enhancement, instruments, etc., can solve the problem of long calculation time, time domain and air domain do not meet the condition of optical flow continuity, and the inability to describe the overall motion displacement of UAV video frames etc. to achieve the effect of reducing computational complexity and controlling computational time

Active Publication Date: 2020-06-09
中国航天电子技术研究院
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For the video images obtained by the payload carried on the UAV, the continuity condition of optical flow is often not satisfied in time domain and air domain
Feature matching can only obtain the optical flow corresponding to the feature points in the local area, and cannot describe the overall motion displacement of the UAV video frame, and the calculation time is long

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
  • A Fast Automatic Target Detection Method
  • A Fast Automatic Target Detection Method
  • A Fast Automatic Target Detection Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] A fast automatic target intrusion detection method, the method is used for target detection of unmanned aerial vehicles, and the method performs Gaussian pyramid layering on the original images in the original video captured by the onboard camera of the unmanned aerial vehicle, to reduce the feature point extraction Computational complexity; Then extract image SIFT feature points for image registration, use pyramid LK sparse optical flow to capture motion information in the image to realize target point motion calculation, move point clustering and eliminate false targets, and finally perform target judgment to achieve Target Detection.

[0046] Such as figure 1 As shown, the method includes the following steps:

[0047] (1) Acquisition of original video: According to the inspection method of the drone on the monitoring area, set the position of the on-board camera to obtain the original video;

[0048] Such as figure 2 As shown, the inspection method includes two m...

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 mainly belongs to the technical field of target intrusion detection, and in particular relates to a target detection method for area intrusion based on unmanned aerial vehicle images. The method is used for the target detection of the unmanned aerial vehicle, and the original image in the original video obtained by the onboard camera of the unmanned aerial vehicle is carried out to Gaussian pyramid layering, to reduce the computational complexity of feature point extraction; then extract the image SIFT Image registration is performed on feature points, and the motion information in the image is captured by the LK sparse optical flow of the pyramid to realize the motion calculation of the target point, the clustering of the motion point and the elimination of false targets, and finally the target judgment to realize the target detection. The method can reduce the search range of inter-frame feature points of the UAV video, overcome the problem of large motion displacement in the UAV image, and improve the detection ability; thereby reducing the intensity of manual detection of area intrusion and improving the automatic perception of the UAV ability.

Description

technical field [0001] The invention mainly belongs to the technical field of target intrusion detection, and in particular relates to a target detection method for area intrusion based on unmanned aerial vehicle images. Background technique [0002] Target detection is the operation of separating the target of interest from the background area of ​​a single frame image or sequence image in a surveillance scene, and identifying and extracting meaningful object entities from the image. The prerequisite for UAVs to complete various tasks is to quickly and accurately detect targets in the surveillance scene. The current research on UAV moving target detection algorithms is in the stage of designing specific methods for specific problems, and the adaptability to complex and changeable working scenarios is poor. Also, the level of object detection required varies depending on the application environment. Generally speaking, the primary task of target detection is to search a ce...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/46G06K9/62G06T7/33G06T7/246
CPCG06T2207/10016G06T2207/20016G06V20/42G06V10/462G06V2201/07G06F18/23
Inventor 黄蜀玲张国勇张杰王静任威许克鹏姜航
Owner 中国航天电子技术研究院
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More