Geometric constraint-based vehicle fine-grained identification method

A technology of geometric constraints and recognition methods, applied in the field of intelligent transportation, can solve the problems of a large number of labeled samples and the inability to further distinguish vehicle types, etc., to meet the needs of obtaining traffic parameters, high stability and precision, and ensure universality.

Inactive Publication Date: 2020-04-14
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 fine-grained vehicle recognition method based on geometric constraints, which overcomes the defects that the existing vehicle recognition methods require a large number of labeled samples and cannot further distinguish vehicle types.

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
  • Geometric constraint-based vehicle fine-grained identification method
  • Geometric constraint-based vehicle fine-grained identification method
  • Geometric constraint-based vehicle fine-grained identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] Specific embodiments of the present invention will be described in detail below. It should be understood that the specific embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0039] The present invention uses a deep learning method to identify vehicle targets, and at the same time utilizes camera calibration results to construct geometric constraints on the vehicle targets to solve their three-dimensional information, and to perform fine-grained vehicle identification on the basis of the three-dimensional information. The method based on a single vanishing point is used for calibration. In the actual road scene, the geometric constraints of the vehicle target are constructed from the two-dimensional target detection results and the calibration results. These parameters are easy to obtain in the road environment, which fully guarantees the generality of this method in this scenario...

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 geometric constraint-based vehicle fine-grained identification method. In a continuous video frame, a deep learning method is utilized to identify a vehicle target. The method comprises the following steps: acquiring a two-dimensional envelope frame of the vehicle target in an image coordinate system and coordinate information of the two-dimensional envelope frame, constructing geometric constraints for the specific vehicle target according to the coordinate information of the two-dimensional envelope frame model in combination with the camera calibration result so asto solve three-dimensional information of the vehicle target, and completing fine-grained identification of the vehicle target on the basis of the three-dimensional information. The method can adaptto different road traffic scenes and uses the camera to extract a large number of vehicle targets in the scene to complete the fine-grained identification process. The method is easy to implement andgood in universality and can be applied to fine-grained identification in various road scenes and is relatively accurate in result.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a vehicle fine-grained recognition method based on geometric constraints. Background technique [0002] Vehicle identification is a very important sub-topic in the intelligent transportation system. It can be applied to the vehicle toll management system and the vehicle inspection system of the public security department. It is one of the key traffic parameters. Due to the continuous increase in the number of vehicles and the large number of similar vehicles on the road, the information that traditional vehicle identification can provide cannot meet the requirements of the management department. Therefore, fine-grained vehicle identification is required to identify more types of valuable vehicle information. [0003] The so-called vehicle fine-grained recognition means that on the basis of identifying the vehicle category (car, truck, etc.), it furt...

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): G06K9/00G06T7/80
CPCG06T7/80G06V20/64G06V20/41G06V2201/08
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段娇娇罗静静杨腾腾蒋维何健乐
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