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

License plate positioning method based on deep learning SSD framework

A license plate positioning and deep learning technology, applied in the field of computer vision recognition, can solve problems such as large amount of calculation and difficulty in achieving real-time effects, and achieve the effect of increasing speed

Active Publication Date: 2018-01-09
JINAN JOVISION TECH CO LTD
View PDF5 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method uses multiple scales and multiple scale reference frames during the training of the RPN convolutional neural network, which can effectively improve the license plate detection of unconventional scales and scales. However, the method is divided into two stages as a whole. First, the rough selection area is obtained. Then classify and regress each roughly selected area, the amount of calculation is still large, and it is difficult to achieve real-time results

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 positioning method based on deep learning SSD framework
  • License plate positioning method based on deep learning SSD framework
  • License plate positioning method based on deep learning SSD framework

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] A license plate location method based on deep learning SSD framework, comprising the following steps:

[0026] 101. Create a license plate dataset based on the VOC dataset format

[0027] First, establish the folder LPdetection for storing the data set, and generate three folders under the LPdetection folder, which are Annotations, ImageSets, and JPEGImages. Name the license plate data file name from 000001.jpg to this type of format, and store the data in the JPEGImages folder. Name the license plate annotation file uniformly starting from 000001.xml, and store it in the Annotations folder. Utilize the existing image data to generate a training sample set and a test sample set, write the image number into trainval.txt and test.txt respectively, and store it in the Main folder under the ImageSets folder. In this embodiment, the data is stored The structure diagram is attached figure 2 shown.

[0028] 102. Convert the dataset to lmdb format

[0029] Specifically, o...

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 positioning method based on deep learning SSD framework, comprising the steps of producing license plate data set based on VOC data set format; transferring thelicense plate data set into 1mdb format; adding an assistant network configuration as SSD framework feature extraction layer and classification layer with ResNet residual error network being the basic network; establishing a SSD framework, and conducting training on the network model by means of the established SSD framework; conducting license plate positioning and model assessment by means of the trained model. The invention is advantageous in that license plate positioning is high in accuracy, low in miss rate, and fast in positioning speed; real-time detection of vehicles can be realized.

Description

technical field [0001] The invention relates to the technical field of computer vision recognition, in particular to a license plate location method based on a deep learning SSD framework. Background technique [0002] Facing the development trend of globalization and informatization in today's world, intelligent transportation system will be the inevitable choice for the development of transportation industry. Through the effective integration and application of advanced information technology, communication technology, control technology, sensor technology, computer technology and system synthesis technology, the interaction between people, vehicles and roads can be presented in a new way, thereby realizing real-time , Accurate, efficient, safe and energy-saving goals. License plate recognition is an important part of modern intelligent transportation systems. License plate recognition technology is applied to road traffic monitoring, automatic recording of traffic violat...

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): G06K9/32G06K9/00G06K9/62G06N3/08
Inventor 闫晓葳房桦韩哲刘琛尹萍
Owner JINAN JOVISION TECH CO LTD
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