Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for recovering license plate image for LPR

A license plate image and image technology, applied in the field of image processing, can solve problems such as difficult identification, achieve accurate identification, enhance robustness, and optimize recovery quality.

Pending Publication Date: 2020-08-25
FOSHAN NANHAI GUANGDONG TECH UNIV CNC EQUIP COOP INNOVATION INST +1
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As for low-quality images captured in harsh environments (such as strong light, night, smog, etc.), or captured under motion, blurred, tilted, etc., it is still difficult to identify

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
  • Method for recovering license plate image for LPR
  • Method for recovering license plate image for LPR
  • Method for recovering license plate image for LPR

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Such as Figure 1 to Figure 3 As shown, the present embodiment discloses a method for recovering a license plate image for LPR, the method mainly includes the following specific steps:

[0036] Step S1: After a series of operations are performed on the images in the known dataset, the images are proportionally divided into training set, verification set and test set.

[0037] Specifically, the S1 step also includes the following steps:

[0038] Step S11: Using several well-known license plate recognition data sets VTLP, divide the training set, verification set and test set according to the ratio of 6:2:2.

[0039] Step S12: In order to increase the amount of training data, the training set is rotated with different angles to generate four sub-pictures, and doubled by size transformation and segmentation methods; the original training picture is marked as I H , and the four rotated subgraphs are denoted as i∈{-30°,-15°,+15°,+30°}, the sub-graph after size transforma...

Embodiment 2

[0055] refer to Figure 1 to Figure 3 , the present embodiment discloses a method for recovering a license plate image for LPR comprising the following steps:

[0056] S1: After a series of operations such as averaging, defogging, and cropping are performed on the images in the known data set, the size of the generated image is 572*572. The image is divided into training set, verification set and test set in proportion.

[0057] The specific steps include the following:

[0058] S11: Use the data set VTLP, which contains 10650 license plate pictures, and then divide it into training set, verification set and test set according to the ratio of 6:2:2, including 6390, 2130 and 2130 license plate pictures respectively.

[0059] S12: Then expand the training set with the correct label, a label map I H Expand into four subimages by rotation transformation i∈{-30°,-15°,+15°,+30°}, four sub-images are generated after size transformation The subgraph after binary segmentation is ...

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 method for recovering a license plate image for LPR, and the method comprises the steps: carrying out a series of operation of an image in a known data set, and dividing theimage into a training set, a verification set and a test set in proportion; establishing a model of an image recovery network for license plate recognition, and training the model by using the training set to obtain a corresponding training model; using the verification set to check the accuracy of the training model, adjusting hyper-parameters of the model, and optimizing the model to obtain better performance; and inputting the test set image into the determined optimization model, testing the generalization performance of the optimization model, and observing how the recovery effect of thelicense plate image is. According to the scheme, the structure of the license plate image restoration network is redesigned, and the auxiliary network is added to optimize the restoration quality of the image, so that the robustness of the LPR is remarkably improved; in addition, a good effect is obtained through a method of combining denoising and a correction network, so that the accuracy of license plate recognition is quite high, and the license plate recognition method is a fast and accurate recognition network.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for restoring license plate images used in LPR. Background technique [0002] With the continuous improvement of the country's economic strength, people's living standards have also been greatly improved, and more and more families have their own private cars. The subsequent traffic problems are also increasing day by day. Today, the problem of traffic vehicle management is one of the important problems in urban management. For this reason, Intelligent Transportation System (Intelligent Transportation System, hereinafter referred to as ITS) came into being. [0003] ITS is the combination of existing science and technology (such as computer counting, sensor technology, image processing technology, etc.) for transportation, service control, etc., and license plate recognition (LPR) is one of the important basic link. At present, the commonly used method is to ...

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): G06T5/00G06K9/34G06K9/62
CPCG06V10/267G06V20/625G06F18/214G06T5/00G06T5/73
Inventor 杨海东陈俊杰黄坤山彭文瑜林玉山
Owner FOSHAN NANHAI GUANGDONG TECH UNIV CNC EQUIP COOP INNOVATION INST