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

An end-to-end lightweight license plate recognition method and device

A license plate recognition, lightweight technology, applied in the field of image processing, can solve the problems of occupying a large amount of memory and low license plate recognition speed

Active Publication Date: 2020-07-24
BEIJING UNIV OF POSTS & TELECOMM
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Because a large number of redundant calculation variables occupy a large amount of memory, and there are redundant calculations, the speed of license plate recognition is low

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
  • An end-to-end lightweight license plate recognition method and device
  • An end-to-end lightweight license plate recognition method and device
  • An end-to-end lightweight license plate recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0173] As an implementation manner of the embodiment of the present invention, the license plate recognition module 520 may include:

[0174] super-resolution image generation unit ( Figure 9 Not shown in), for inputting the candidate frame coordinates and the feature map into the target super-resolution generation network; from the region corresponding to the candidate frame coordinates in the feature map, extract the license plate feature; according to The extracted license plate features are used to generate super-resolution license plate images.

[0175] As an implementation manner of an embodiment of the present invention, the recognition model training module ( Figure 9 not shown) may include:

[0176] Identify model building blocks ( Figure 9 Not shown in), for building the initial license plate recognition model, wherein, the initial license plate recognition model includes an initial feature extraction network, an initial region candidate location network, an in...

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

An embodiment of the present invention provides an end-to-end lightweight license plate recognition method and device, the method comprising: acquiring an image to be recognized; and obtaining the license plate in the image to be recognized based on the image to be recognized and a pre-trained target license plate recognition model The number and its position coordinates in the image to be recognized, wherein the target license plate recognition model includes a target feature extraction network, a target area candidate positioning network, a target super-resolution generation network and a target recurrent neural network. Since in this scheme, the image to be recognized is input into the target license plate recognition model, the target license plate recognition model can output the number of the license plate and its position coordinates in the image to be recognized, realizing an end-to-end model, and at the same time, calculation variables such as image features can be Reuse, no need for repeated calculations, the model occupies less memory, and the speed of license plate recognition is greatly improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an end-to-end lightweight license plate recognition method and device. Background technique [0002] License plate recognition technology is the foundation of intelligent transportation. In recent years, with the maturity of digital image processing, pattern recognition and artificial intelligence technology, license plate recognition technology is also developing continuously. [0003] The general license plate recognition process includes three steps: license plate location, character segmentation, and character recognition. These three steps are realized through multiple deep learning models. For example, the license plate location generally realizes the feature extraction of the image to be recognized through the feature extraction model, obtains the feature map, and determines the position of the candidate frame through the region candidate location model. The char...

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 Patents(China)
IPC IPC(8): G06K9/32G06T3/40G06V10/764
CPCG06T3/4046G06V10/255G06V20/625G06T3/4053G06N3/084G06V20/63G06V10/82G06V10/764G06N3/044G06V20/584G06F18/24133
Inventor 傅慧源马华东张逸凡曹雨张登奎
Owner BEIJING UNIV OF POSTS & TELECOMM