License plate recognition model training method and device and license plate recognition method and device

A license plate recognition and model training technology, applied in the field of license plate recognition, can solve problems such as recognition and detection accuracy, achieve the effect of improving efficiency, meeting the requirements of accuracy, and simplifying the process of license plate recognition

Active Publication Date: 2021-02-05
ZHUHAI DAHENGQIN TECH DEV CO LTD
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Therefore, the technical problem to be solved in the present invention is to overcome the defect that the license plate detection and recognition model in the prior art cannot perform license plate recognition detection on vehicles with double license plates, and thus provide a license plate recognition model training method

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
  • License plate recognition model training method and device and license plate recognition method and device
  • License plate recognition model training method and device and license plate recognition method and device
  • License plate recognition model training method and device and license plate recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] Embodiment one, such as figure 1 A kind of license plate recognition model training method shown in -6, comprises the following steps:

[0060] Step S10, constructing a double license plate vehicle data set B and a license plate data set C according to the collected vehicle image samples with double license plates;

[0061] Step S20, training the yolov3-tiny network model structure based on the double license plate vehicle data set B, and locating the license plate area in the vehicle picture;

[0062] Step S30, perform a network test on each training model of yolov3-tiny according to the double-license plate vehicle test set containing a variety of test pictures with double license plates, and retain the weight file of the yolov3-tiny network model with the highest accuracy;

[0063] Step S40, training the network model structure of LPS / CR-NET character segmentation and recognition based on the license plate data set C, locating the position of each character in the l...

Embodiment 2

[0120] Embodiment 2, a method for license plate recognition, comprising the following steps:

[0121] 1) Obtain vehicle pictures;

[0122] 2) input the vehicle picture into the model, carry out double license plate recognition to the vehicle picture, and obtain the double license plate recognition result; wherein, the model is obtained by training the license plate recognition model training method described in the above-mentioned implementation one;

[0123] 3) Using the character sorting algorithm to the double license plate recognition results obtained in step 2)

[0124] 4) Output two license plate numbers.

Embodiment 3

[0125] Embodiment 3, a double license plate recognition device, comprising:

[0126] The acquisition module is used to acquire vehicle pictures;

[0127] The recognition module is configured to input the vehicle picture into a model, and perform double license plate recognition on the vehicle picture, wherein the model is obtained by training the license plate recognition model training method described in the first embodiment above.

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 license plate recognition model training method and device and a license plate recognition method and device. The model training method comprises the steps: constructing a double-license-plate vehicle data set and a license plate data set; based on the double-license-plate vehicle data set, training a yolov3-tiny network model; testing the training model of the yov3-tiny,and reserving the weight of the yov3-tiny network model with the highest precision; training an LPS / CR-NET network model based on the license plate data set, carrying out network testing on the LPS / CR-NET network model, and preliminarily determining several network model weights with good precision; and calculating the double-license-plate recognition accuracy under each preliminarily determinedLPS / CR-NET network model weight on the basis of a double-license-plate detection result of yoov3-tiny and a character sorting method, and retaining the LPS / CR-NET network model weight with the highestaccuracy. A quick and efficient license plate positioning method is provided for double-license-plate recognition, and the double-license-plate recognition precision is effectively improved.

Description

technical field [0001] The invention relates to the technical field of license plate recognition, in particular to a deep learning-based license plate recognition model training method, a license plate recognition method and a device. Background technique [0002] Vehicles passing through the electronic fence area not only have the diversity of vehicle license plate formats in my country (ordinary car, embassy car, police car, military vehicle, and Hong Kong and Macao license plate format differences) and the difficulty of Chinese character recognition, but also consider the different license plate formats of other countries. format, and different vehicle license plate suspension positions are also different, which puts forward new requirements for license plate recognition technology. The license plate recognition technology requires the ability to extract and recognize the moving vehicle license plate from the complex background, and identify the vehicle plate number, color...

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): G06K9/62G06K9/34G06K9/32G06N3/04
CPCG06V10/243G06V30/153G06V20/625G06V30/10G06N3/045G06F18/214
Inventor 邓练兵李大铭李皓
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

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