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

Intelligent license plate recognition method and system based on deep learning

A technology of intelligent license plate and deep learning, applied in neural learning methods, character and pattern recognition, image enhancement, etc., can solve the problems of slow response speed and low precision of automatic license plate recognition, and achieve high average recognition accuracy and enhanced contrast Effect

Pending Publication Date: 2022-04-12
NANJING COMM INST OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] What the present invention needs to solve is the technical problem of slow response speed and low precision of license plate automatic recognition existing in the prior art

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
  • Intelligent license plate recognition method and system based on deep learning
  • Intelligent license plate recognition method and system based on deep learning
  • Intelligent license plate recognition method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Such as figure 1 As shown, the embodiment of the present invention provides a kind of intelligent license plate recognition method based on deep learning, comprising the following steps:

[0045]S1, obtain the original vehicle picture, use CenterNet to locate, detect and save the target position of the license plate on the original vehicle picture; use the camera preset at the access control to obtain the original vehicle picture of each passing vehicle, and then perform license plate recognition and detection , the general process of license plate recognition includes three steps: acquisition of original vehicle pictures, license plate detection and license plate recognition.

[0046] S2, extract the license plate interest area image, input it into the trained license plate recognition model, judge whether the license plate is obscured, if normal, further extract the license plate image, if not normal, prompt an alarm. Wherein, the ROI image of the license plate may b...

Embodiment 2

[0079] Compared with Embodiment 1, the only difference is that S5 is also included to obtain the original vehicle picture and license plate image information, then remove the overlapping part of the original vehicle picture and license plate image information to obtain background image information containing blank parts, and convert the background image information to Perform grayscale and binarization processing, take the central axis of the blank part as the starting line, and search for axisymmetric graphics on the left and right sides perpendicular to the starting line, if the area value of the axisymmetric graphics is greater than the preset value, then The license plate image information is on the vehicle; if the area of ​​the axisymmetric figure is smaller than the preset value, the original vehicle picture and environmental sound information are obtained again, and the original vehicle picture is grayscaled and binarized, and then judged to be in the blank Whether there...

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 belongs to the technical field of intelligent traffic engineering, and particularly provides an intelligent license plate recognition method and system based on deep learning, and the method comprises the steps: obtaining an original vehicle picture, and carrying out the positioning, detection and storage of a target position of a license plate of the original vehicle picture through employing Center Net; extracting a license plate region-of-interest image, inputting the license plate region-of-interest image into the trained license plate recognition model, judging whether the license plate is unclearly shielded, if so, further extracting a license plate picture, and if not, prompting to give an alarm; preprocessing the license plate picture to obtain license plate image information; and transmitting the processed license plate characters into a convolutional neural network for model training and optimization, and finally predicting to obtain a license plate recognition result. The Center Net network is selected to carry out license plate positioning and detection, and image preprocessing technologies such as histogram equalization and inclination correction are selected to enhance the contrast of the license plate, so that an effective guarantee is provided for subsequent training of the convolutional neural network. The license plate information can be effectively distinguished, and the average recognition accuracy is high.

Description

technical field [0001] The present invention relates to the technical field of intelligent traffic engineering, and more specifically, to a method and system for intelligent license plate recognition based on deep learning. Background technique [0002] Intelligent license plate recognition is an important part of modern intelligent transportation systems and is widely used in highways, parking lots, intersections and other scenarios. With the continuous development of big data and artificial intelligence, intelligent license plate recognition has greatly improved in data processing, adaptive learning and special scene training, and has stronger fault tolerance and robustness. Through the automatic identification and tracking of license plate numbers, it can effectively reduce the cost of vehicle automation management, regulate irregular vehicle behavior, and provide a solid guarantee for social stability and residents' convenient life. [0003] The general process of licen...

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): G06V20/54G06V10/25G06T7/11G06T5/40G06T3/40G06T3/00G06N3/08G06N3/04
CPCY02T10/40
Inventor 季丹
Owner NANJING COMM INST OF TECH
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