Vehicle spatial form recognition method and system based on contour constraint

A technology of spatial form and identification method, applied in the field of intelligent transportation, can solve the problems of high cost and inability to further identify the three-dimensional information of vehicles, and achieve the effect of high stability and accuracy, wide application and universality.

Pending Publication Date: 2020-07-31
CHANGAN UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects and deficiencies in the prior art, the present invention provides a vehicle space form recognition method and system based on contour constraints, which overcomes the defects of the existing vehicle space form recognition method, such as high cost and inability to further identify three-dimensional information of the vehicle.

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
  • Vehicle spatial form recognition method and system based on contour constraint
  • Vehicle spatial form recognition method and system based on contour constraint
  • Vehicle spatial form recognition method and system based on contour constraint

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Such as Figure 1 to Figure 8 As shown, the present invention discloses a vehicle space shape recognition method based on contour constraints, and the detailed steps are as follows:

[0044] Step 1, input the traffic video, obtain the video image of each frame, and form an image sequence;

[0045] Step 2, extract the vanishing point from the video image obtained in step 1, establish the camera model and coordinate system, the two-dimensional envelope model and the three-dimensional envelope frame model of the vehicle target in the image coordinate system, and perform camera calibration in combination with the vanishing point, Obtain camera calibration parameters and horizon information;

[0046] Refer to the method of the paper "A Taxonomy and Analysis of Camera Calibration Methods for Traffic Monitoring Applications", such as figure 2 As shown, establish the camera model, world coordinate system O-XYZ, camera coordinate system O-X C Y C Z C , The image coordinate...

Embodiment 2

[0078] This embodiment provides a vehicle space shape recognition system based on contour constraints, the system includes:

[0079] The data input module is used to input the traffic video, obtains the video image of each frame, and forms an image sequence;

[0080] The camera calibration module is used to establish the camera model and coordinate system, the two-dimensional envelope model and the three-dimensional envelope frame model of the vehicle target in the image coordinate system, perform camera calibration, and obtain the internal and external parameters of the camera and the horizon information of the scene;

[0081] The vehicle target detection and segmentation module is used to detect and segment the vehicle target in the video image using the method of deep learning Mask RCNN. The detection result includes the two-dimensional coordinate information of the vehicle target under the two-dimensional envelope model and the vehicle target The vehicle category of , the ...

Embodiment 3

[0085] In order to verify the validity of the method proposed by the present invention, an embodiment of the present invention adopts the following Figure 4 Shown is an image of an actual road traffic scene in which a single vanishing point along the road direction is identified and the camera is calibrated. Such as Figure 6 As shown, on this basis, the vehicle target is detected and segmented by the method of deep learning Mask RCNN, and the coordinates of the three-dimensional envelope reference point of the vehicle target in the image coordinate system are obtained. Such as Figure 7 Shown is a schematic diagram of constructing contour constraints, combined with the calibration results, constructing contour constraints and solving them. Such as Figure 8 Shown is the result map of vehicle space shape recognition.

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 vehicle spatial form recognition method and system based on contour constraint, and the method comprises the steps: recognizing a vehicle target in a continuous image sequence through employing a deep learning Mask RCNN method, and obtaining a two-dimensional envelope box and segmented image information of the vehicle target under an image coordinate system; calculating acontour of the segmented image of the vehicle, obtaining a contour point set of each vehicle target, and calculating a contour gravity center point of the contour point set; and then, according to the coordinate information of the two-dimensional enveloping frame model, solving a convex hull of each vehicle target under the three-dimensional enveloping frame model in combination with a camera calibration result and horizon information, and constructing a contour constraint for a specific vehicle target, thereby solving spatial form information of the vehicle target. The method can adapt to different road traffic scenes, and uses the camera to extract a large number of vehicle targets in the scene to complete the spatial form recognition process. The method can be applied to vehicle spaceform identification in various road scenes, and is accurate in result, simple to implement and good in universality.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a vehicle space shape recognition method and system based on contour constraints. Background technique [0002] Vehicle spatial shape recognition has a wide range of applications in the field of autonomous driving. Through the perception of the surrounding environment, the vehicle's pose state information (including the vehicle's three-dimensional size, spatial position, and motion direction, etc.) can be obtained in real time, and the safety status of the vehicle can be analyzed to provide The path planning and obstacle avoidance of the vehicle provide the basis, enabling the self-driving vehicle to make effective decisions and ensure that the vehicle reaches its destination safely. [0003] Commonly used vehicle pose estimation methods mainly rely on sensor fusion methods, such as laser sensors, GPS, inertial navigation systems, etc. Although thes...

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): G06T7/11G06T7/64G06T7/80G06T17/00G06K9/00G06K9/46G06N3/00G06N3/04G06N3/08
CPCG06T7/11G06T7/80G06T7/64G06T17/00G06N3/006G06N3/08G06V20/64G06V20/584G06V10/44G06V2201/08G06N3/045Y02T10/40
Inventor 王伟唐心瑶宋焕生张朝阳梁浩翔张文涛戴喆云旭侯景严刘莅辰贾金明李俊彦武非凡雷琪杨露余宵雨靳静玺王滢暄赵锋穆勃辰李聪亮
Owner CHANGAN UNIV
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