Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Coarse-to-fine two-stage unconstrained license plate region accurate extraction method

A technology of license plate area and extraction method, applied in the field of license plate information recognition, can solve the problems of multiple redundant information, affecting the accuracy of license plate character recognition, etc., and achieve the effect of good input, expanding the scope of the target area, and improving the detection accuracy.

Pending Publication Date: 2020-10-30
湖北九感科技有限公司 +1
View PDF9 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The license plate images obtained in real scenes usually have different degrees of inclination. For the license plate with a large inclination angle, the license plate area positioned with a rectangular frame often contains more redundant information, which affects the accuracy of license plate character recognition.

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
  • Coarse-to-fine two-stage unconstrained license plate region accurate extraction method
  • Coarse-to-fine two-stage unconstrained license plate region accurate extraction method
  • Coarse-to-fine two-stage unconstrained license plate region accurate extraction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] By changing the MobileNet network structure and combining the powerful target detection capability of YOLOv3 while giving full play to its advantages of fast operation speed, the present invention proposes a two-stage unconstrained accurate license plate area extraction method from coarse to fine. YOLOv3 is used for rough positioning of the license plate area, and MobileNet extracts the vertices of the license plate in the rough positioning area. Due to the small size of the processed image, the MobileNet network can still achieve good results. The method proposed by the invention can accurately locate the vertices of the license plate in an unconstrained environment, and the detection speed is fast. Figure 1(1)~Figure 1(8) As shown, among them, the green rectangular frame in Figure 1 (1) to Figure (4) is the license plate positioning effect based on the rectangular frame, the red point in Figure (5) to Figure (8) is the vertex position of the license plate, and the yel...

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 coarse-to-fine two-stage unconstrained license plate region accurate extraction method. The method comprises the steps: employing a YOLOv3 network for carrying out coarse positioning on a license plate area in a license plate image, and then employing an improved network based on MobileNetV3 for determining the vertex position of a license plate in the coarse positioningarea, and finally employing the perspective transformation for correcting the finely positioned license plate area into a rectangular area. Compared with a one-stage network for directly carrying outlicense plate positioning, the two-stage license plate positioning method has more obvious advantages in the aspect of license plate positioning precision, and meanwhile, the real-time performance oflicense plate positioning is also maintained.

Description

technical field [0001] The invention relates to the technical field of license plate information recognition, in particular to a two-stage unconstrained license plate area precision extraction method from coarse to fine. Background technique [0002] The license plate recognition system is a vital part of the intelligent transportation system, and its application scenarios are extensive, such as high-speed toll stations, parking lot entrances and exits, traffic violation monitoring, traffic law enforcement, etc. In road traffic management, the license plate is the only basis for obtaining vehicle owner information, so it is of great significance to accurately and effectively recognize the license plate information during the driving process of the vehicle. As a key part of automatic license plate recognition system, license plate detection has always been a research topic that has attracted much attention. Existing license plate detection techniques can be roughly divided i...

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): G06K9/32G06K9/62G06N3/04
CPCG06V10/242G06V20/63G06V20/625G06N3/045G06F18/24G06F18/214
Inventor 徐光柱匡婉雷帮军万秋波吴正平石勇涛
Owner 湖北九感科技有限公司
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
Eureka Blog
Learn More
PatSnap group products