Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

License plate recognition method and device for intelligent city

A license plate recognition and urban technology, applied in the field of artificial intelligence, can solve the problems of low resolution and low accuracy of low resolution license plate recognition, and achieve good recognition effect

Pending Publication Date: 2022-02-11
李显德
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In some scenarios, the vehicle is far away from the monitoring probe, and the resolution of the license plate in the collected image is relatively low. However, the existing technology has a low accuracy rate for low-resolution license plate recognition and needs further improvement.

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 method and device for intelligent city
  • License plate recognition method and device for intelligent city
  • License plate recognition method and device for intelligent city

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] With the help of deep learning framework, according to figure 1 The structure shown is used to build a classification network for the license plate image 1. In this embodiment, six GMT feature extraction mechanisms 22 and five residual feature extraction units 4 are set. Both the head convolutional layer 21 and the middle convolutional layer 8 are convolutional layers with a convolution kernel size of 3*3, and the structure of the fusion module 25 is as follows figure 2 As shown, the internal structure of the GMT feature extraction mechanism 22 is as follows image 3 As shown, the internal structure of the channel attention module 26 is as follows Figure 4 As shown, the internal structure of the spatial attention module 27 is as follows Figure 5 As shown, the image enlargement unit 24 is realized by using existing technology, and its internal structure is as follows Figure 6 As shown, it includes a sequentially connected 3*3 convolutional layer, a sub-pixel convo...

Embodiment 2

[0053] In order to reflect the functions of the channel attention module 26 and the spatial attention module 27, a comparative experiment is set up in Embodiment 2. On the basis of Embodiment 1, the spatial attention module 27 in the GMT feature extraction mechanism 22 is removed separately, as Embodiment 2A. On the basis of Embodiment 1, the spatial attention module 27 and the channel attention module 26 in the GMT feature extraction mechanism 22 are removed, as Embodiment 2B. To ensure that the implementation process, data set, batch-size and epoch are exactly the same, the comparative experiment is carried out. The experimental results are shown in the following table:

[0054]

[0055] It can be seen from the above results that setting the spatial attention module 27 and the channel attention module 26 both help to improve the recognition accuracy of low-resolution license plates.

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 method and device for a intelligent city. The recognition method comprises the steps of obtaining a segmented license plate image, obtaining a trained license plate image classification network, performing super-resolution reconstruction on the license plate image by using a reconstruction sub-network, classifying the super-resolution license plate image by using a classification sub-network and the like. The classification subnet comprises a middle convolutional layer, a residual feature extraction unit, a tail global average pooling layer, a tail full connection layer and a tail softmax activation layer, and the reconstruction subnet comprises a head convolutional layer, a GMT feature extraction mechanism, a feature fusion unit and an image amplification unit. According to the method, the reconstruction subnet and the classification subnet are connected together to form a network, the image output after reconstruction of the reconstruction subnet is most beneficial to low-resolution license plate recognition, the network is high in low-resolution license plate recognition accuracy, the operation speed after deployment is higher than that of a conventional two-step method, and the construction requirement of a intelligent city can be better met.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence, and in particular relates to a license plate recognition method and device for smart cities. Background technique [0002] With the development of artificial intelligence, cloud computing, Internet of Things and new-generation information technology, smart cities have gradually transformed from concept to reality, providing residents with more convenient and complete infrastructure. Relying on new technological means to make urban management intelligent, further mobilize the enthusiasm of the government, enterprises and citizens, and enhance people's sense of security and happiness. [0003] Smart transportation is an important part of smart cities. At present, a large number of monitoring probes have been deployed on the roads in my country, which can quickly obtain information such as license plates, vehicle models, and driver status, and supervise and regulate traffic behavior....

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
IPC IPC(8): G06V20/62G06V30/148G06V10/44G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08G06T3/40
CPCG06N3/08G06T3/4053G06N3/047G06N3/048G06N3/045G06F18/2415G06F18/241
Inventor 李显德
Owner 李显德
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com