Method and device for locating license plate based on fully convolutional network

A fully convolutional network and license plate positioning technology, which is applied in the field of license plate positioning methods and devices based on fully convolutional networks, can solve problems such as multiple vehicles, low resolution, and low brightness, and achieve the effect of improving positioning efficiency and accuracy.

Active Publication Date: 2017-07-21
张东森
View PDF3 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the research has made great progress in recent years, it still cannot solve the problem of license plate location in natural scenes, low brightness, low resolution, multiple vehicles, and vehicle tilting.

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 and device for locating license plate based on fully convolutional network
  • Method and device for locating license plate based on fully convolutional network
  • Method and device for locating license plate based on fully convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0020] figure 1 It is a flow chart of a method for locating a license plate based on a fully convolutional network provided in Embodiment 1 of the present invention. The method of this embodiment can be executed by a license plate location device based on a fully convolutional network, and the device can be implemented by means of hardware and / or software. refer to figure 1 The license plate location method based on the full convolutional network provided in this embodiment may specifically include the following:

[0021] Step 11: Using the license plate location model trained in advance based on the first fully convolutional neural network structure, determine at least one preliminary area of ​​the license plate image included in the vehicle image to be detected.

[0022] Among them, the first fully convolutional neural network structure can be a self-defined FCN8 fully convolutional network, for example, it can be obtained by modifying the fully connected layer structure a...

Embodiment 2

[0049] figure 2 It is a flow chart of a method for locating a license plate based on a fully convolutional network provided in Embodiment 2 of the present invention. refer to figure 2 , the method may specifically include:

[0050] Step 21. Using the license plate location model trained in advance based on the first fully convolutional neural network structure, determine at least one preliminary area of ​​the license plate image contained in the vehicle image to be detected.

[0051] Step 21: Process the image of the vehicle to be detected including the at least one pre-divided area of ​​the license plate image to obtain the candidate license plate area contained in the image of the vehicle to be detected.

[0052] Step 23: Using the license plate initial character recognition model trained in advance based on the second fully convolutional neural network structure, determine whether the initial character of the candidate license plate area is the license plate initial cha...

Embodiment 3

[0060] This embodiment provides a license plate location device based on a fully convolutional network. image 3 It is a structural diagram of a license plate location device based on a fully convolutional network provided in Embodiment 3 of the present invention, as shown in image 3 As shown, the license plate location device based on the full convolutional network can include:

[0061] The license plate preliminary division module 31 is used to adopt the license plate location model obtained based on the first full convolutional neural network structure training in advance to determine at least one license plate picture initial division area contained in the vehicle picture to be detected;

[0062] The candidate license plate module 32 is configured to process the vehicle picture to be detected that includes the at least one license plate image preliminary division area, and obtain the candidate license plate area contained in the vehicle picture to be detected;

[0063] T...

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 embodiments of the invention disclose a method and a device for locating a license plate based on a fully convolutional network. The method comprises the following steps: using a license plate location model pre-trained based on a first fully convolutional network structure to determine at least one license plate image preliminary segmentation area contained in a to-be-detected vehicle image; processing the to-be-detected vehicle image containing at least one license plate image preliminary segmentation area to get a candidate license plate area contained in the to-be-detected vehicle image; using a license plate first character recognition model pre-trained based on a second fully convolutional network structure to determine whether the first character in the candidate license plate area is a first character of a license plate; and if the first character in the candidate license plate area is a first character of a license plate, determining that the candidate license plate area is a license plate, and marking the position of the license plate in the to-be-detected vehicle image. According to the embodiments, a license plate area can be located from a to-be-detected vehicle image in a natural scene, and the efficiency and the precision of license plate area locating are improved.

Description

technical field [0001] Embodiments of the present invention relate to the technical fields of computer vision and machine learning, and in particular to a license plate location method and device based on a fully convolutional network. Background technique [0002] At present, in the field of intelligent transportation, license plate location has high research and application value, such as automatic toll collection in parking lots, automatic search for license plate information in traffic monitoring, etc. Although the research has made great progress in recent years, it still cannot solve the problem of license plate location in environments such as natural scenes, low brightness, low resolution, multiple vehicles, and vehicle tilt. Therefore, how to locate the license plate area from the vehicle video image in natural scenes is an important issue in applications such as license plate recognition systems, traffic monitoring, and vehicle access control. Contents of the inv...

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): G06K9/32G06K9/62G06N3/04
CPCG06V10/25G06V20/625G06V30/10G06N3/045G06F18/214
Inventor 向函符祖峰赵勇谢锋陈胜红
Owner 张东森
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