Two-place double-license-plate detection and recognition method and system based on deep learning

A license plate detection and deep learning technology, applied in the field of double license plate detection and recognition in two places, can solve problems such as the effect of license plate positioning effect, and achieve the effect of improving the detection and recognition accuracy, improving the effect, and reducing the time consumption.

Active Publication Date: 2020-09-15
ZHUHAI DAHENGQIN TECH DEV CO LTD
View PDF14 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, when facing double license plates, the size of license plates in mainland China, Hong Kong, and Macau is different, and there are two license plates in a small area, which greatly affects the positioning effect of license plates

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
  • Two-place double-license-plate detection and recognition method and system based on deep learning
  • Two-place double-license-plate detection and recognition method and system based on deep learning
  • Two-place double-license-plate detection and recognition method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0034] The embodiment of the present invention provides a method for detecting and recognizing double license plates in two places based on deep learning. The detection and recognition process is as follows: figure 1 As shown, the following are the training steps and usage steps of each network used in the double license plate detection and recognition method in the two places:

[0035] S1, constructing a large-scale vehicle data set with license plate labeling; specifically includes the following steps:

[0036]S1-1, collect the image data set with the vehicle, and then use the pre-training model of YOLOv3 on the official website to detect the vehicle target on the collected image data set and segment the vehicle area from the original image for storage to obtain the vehicle image data set. YOLOv3 is an existing target detection method, and ...

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 provides a two-place double-license-plate detection and recognition method and system based on deep learning, and the two-place double-license-plate detection and recognition method comprises the steps: constructing a vehicle data set with license plate marks, and carrying out the sample marking; training an improved MTCNN license plate detector, positioning coordinates of a licenseplate area and four corners of a license plate in the image by using the trained license plate detector, and correcting the license plate area by using the coordinates of the four corners of the license plate; training a license plate character detector, and detecting bounding boxes and categories of all characters in the license plate; and sorting the license plate character detection results, and outputting a serial license plate character sequence. The two-place double-license-plate detection and recognition method is mainly used for license plate detection and recognition of vehicles in two places in a complex scene, and the used networks are light networks, and the license plate detection and recognition accuracy can be improved while the time consumed for license plate detection andrecognition is effectively shortened. In addition, four angular point coordinates of the license plate are adopted for license plate correction, and the inclination and distortion conditions of the license plate are effectively improved, and the license plate recognition effect is improved.

Description

technical field [0001] The invention belongs to the technical field of license plate detection, and in particular relates to a technical scheme of a detection and recognition method for dual license plates in two places. Background technique [0002] Vehicles are an important means of transportation, and their management is an important part of urban traffic management and safety management. As the identification code of the vehicle, the license plate can uniquely represent the vehicle object, so the management and identification of vehicles using the license plate has become the hottest and most challenging task in recent years. [0003] The detection and recognition of the license plate is the technology of locating the position of the license plate from the complex natural image and recognizing the license plate number. The process of license plate detection and recognition is generally divided into two categories, one is vehicle detection, license plate location, and li...

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): G06K9/32G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/63G06V30/158G06V30/153G06V20/625G06N3/045G06F18/214
Inventor 邓练兵
Owner ZHUHAI DAHENGQIN TECH DEV CO LTD
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