A three-dimensional moving target detection method and system

A three-dimensional motion and target detection technology, applied in the field of computer vision, can solve the problems of inaccurate detection effect, large amount of data analysis, and high detection false alarm rate, so as to reduce the false alarm rate, reduce the detection area, and improve the detection efficiency.

Active Publication Date: 2019-11-08
湖南峰华智能科技有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The embodiment of the present invention provides a method and system for denoising point cloud data to solve the problem of high false alarm rate and low detection efficiency of the existing feature-based target recognition and detection methods, and the data analysis volume of the method based on scene classification. Large, inaccurate detection problems

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 three-dimensional moving target detection method and system
  • A three-dimensional moving target detection method and system
  • A three-dimensional moving target detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] figure 1 A flow chart of steps of a method for detecting a three-dimensional moving object according to Embodiment 1 of the present invention is shown.

[0034] refer to figure 1 , the three-dimensional moving object detection method of the present embodiment comprises the following steps:

[0035] Step S100, acquiring a 3D range image of a multi-view scene to be processed.

[0036] In this step, the three-dimensional range image includes three-dimensional moving targets and background interference, and the three-dimensional range image can be real data obtained by a distance sensor (such as a laser radar, etc.), or simulated data obtained by a simulation software.

[0037] In order to ensure the speed of transmission and processing, drawing on the parallel processing method of multiple receptive fields of the human retina, the master control computer platform synchronously transmits the 3D distance images of the multi-view scenes to be processed to multiple network c...

Embodiment 2

[0055] figure 2 A flow chart of steps of a method for detecting a three-dimensional moving object according to Embodiment 2 of the present invention is shown.

[0056] refer to figure 2 , the three-dimensional moving object detection method of the present embodiment comprises the following steps:

[0057] Step S200, acquiring real data or simulated data including a three-dimensional range image.

[0058] In order to ensure the speed of transmission and processing, drawing on the parallel processing method of multiple receptive fields of the human retina, the master control computer platform transmits the three-dimensional range images to be processed synchronously to multiple network computers to realize parallel multi-channel processing. The 3D range image usually contains multiple 3D moving targets and background interference. The 3D range image can be the real data obtained by the distance sensor (such as lidar, etc.), or the simulated data obtained by the simulation so...

Embodiment 3

[0087] Figure 4 A structural block diagram of a three-dimensional moving object detection system according to Embodiment 3 of the present invention is shown.

[0088] The 3D moving object detection system in this embodiment includes: an acquisition module 400, configured to acquire a 3D range image of a multi-view scene to be processed, the 3D range image including a 3D moving object; an extraction module 402, configured to extract the multiple The local invariant feature information of the three-dimensional range image of the viewing angle scene, and determine the feature description vector of the local invariant feature information; the registration module 404 is used to align the local invariant feature information and the feature description vector according to the local invariant feature information and the feature description vector The three-dimensional range image of the multi-view scene is registered; the determination module 406 is used to determine the candidate de...

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

An embodiment of the present invention provides a method and system for detecting a three-dimensional moving object, wherein the method includes: acquiring a three-dimensional range image of a multi-view scene to be processed, where the three-dimensional range image includes a three-dimensional moving target; extracting a three-dimensional distance of the multi-view scene The local invariant feature information of the image, and determine the feature description vector of the local invariant feature information; register the 3D range image of the multi-view scene according to the local invariant feature information and feature description vector; according to the registered multi-view scene The local invariant feature information of the 3D range image is used to determine the candidate detection area; the candidate detection area is verified and reviewed by iterative ground estimation and elevation filtering to obtain the accurate detection area; the 3D moving target detection is performed on the accurate detection area. The embodiment of the present invention reduces the detection false alarm rate, improves the detection efficiency, reduces the amount of data processing, and obtains a more accurate detection effect through repeated iterations.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of computer vision, and in particular to a method and system for detecting a three-dimensional moving object. Background technique [0002] Three-dimensional moving target detection in complex ground environment is a new research hotspot and difficulty in the field of computer vision technology. 3D moving target detection is expected to break through the inherent bottlenecks in existing information processing systems based on 1D and 2D images, and explore new research directions for target detection. Due to the influence of noise, background interference, viewing angle and other factors, the detection of 3D moving objects on complex ground has always been one of the bottlenecks restricting the development and application of 3D intelligent information processing. Currently, 3D moving target detection is facing severe challenges, which mainly include: [0003] 1) Background clutter inte...

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): G06T7/246G01S7/48G01S17/89
CPCG01S7/4802G01S17/89
Inventor 滕书华谭志国张鹏
Owner 湖南峰华智能科技有限公司
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