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

A non-binarization and edge detection method for license plate character image segmentation

An edge detection and character image technology, applied in the field of license plate image recognition, can solve the problem of difficult segmentation of license plate characters, and achieve the effect of good generalization ability, improved character/non-character judgment, and strong generalization ability.

Active Publication Date: 2021-05-18
ZHEJIANG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem of difficult segmentation of license plate characters in complex environments in the background technology, the present invention provides a non-binarization and edge detection method for segmenting license plate character images. The recognition method is simple, the recognition accuracy is high, and it has better practicality and scalability

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
  • A non-binarization and edge detection method for license plate character image segmentation
  • A non-binarization and edge detection method for license plate character image segmentation
  • A non-binarization and edge detection method for license plate character image segmentation

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0062] The present invention adopts such as image 3 The picture shown on the left is a character sample picture, using the image 3 The images shown on the right serve as non-character sample images, which include partial or no-character sample images containing only complete characters. Such as Figure 4 and Figure 5 As shown, the license plate characters are segmented in the simple license plate character area and the more complex license plate character area (including complex shading), and the corresponding character probability distribution and character segmentation results after non-maximum suppression are obtained. The image marked by the box The region is each character region after character segmentation in the simple and complex license plate regions respectively. As a comparison, the license plate in this embodiment is binarized and edge detected respectively, and the effect is as follows Figure 6 .

[0063] In traditional image processing, methods such as ...

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 non-binarized and edge-detected license plate character image segmentation method. Obtain the character sample pictures and non-character sample pictures of the license plate; build a deep learning network, input training to obtain the character discrimination network; obtain the license plate picture, detect and locate the character area of ​​the license plate picture to obtain the license plate character area, set the sliding window; set the sliding window A series of windows are formed by sliding and traversing from left to right in the character area, and the image area in the window is input to the character discrimination network to obtain the character probability of whether a single character is contained in the window, and optimize the processing. The reserved window is the license plate. character. The present invention can accurately perform character segmentation on the license plate character area. Compared with the method of character segmentation using binarization and edge detection under traditional image processing, it has excellent segmentation for license plates with shading and uneven illumination. As a result, the invention has great practical value for license plate recognition applications.

Description

technical field [0001] The invention relates to a method for segmenting license plate characters in the field of license plate image recognition, in particular to a non-binarization and edge detection method for segmenting license plate character images. Background technique [0002] As an important information carrier of a vehicle, the license plate provides the only legal representation for the vehicle. License plate recognition technology is also the most core and key technology in today's smart transportation, which plays an important role in timely tracing, querying vehicle information and avoiding certain traffic accidents. [0003] The current license plate recognition technology mainly includes license plate location, license plate character segmentation, license plate character recognition and other steps. Among them, license plate character segmentation is an extremely important and most complicated step in the license plate recognition process. At present, the l...

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): G06K9/32G06K9/34G06K9/62
CPCG06V20/63G06V30/153G06V10/267G06V20/625G06F18/241
Inventor 祁忠琪涂凯
Owner ZHEJIANG UNIV
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