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

Vehicle license plate detection and recognition method based on combination of MSER and SWT

A license plate detection and recognition method technology, applied in the field of vehicle license plate detection and recognition, can solve the problems of mistaken deletion and missed detection of candidate regions, long algorithm time, etc., to solve the sensitivity of blur and threshold, reduce algorithm time, and reduce algorithm time complexity. degree of effect

Inactive Publication Date: 2018-01-09
UNIV OF SHANGHAI FOR SCI & TECH
View PDF4 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This algorithm can effectively detect text under complex backgrounds under certain conditions, but it still has the following disadvantages: ① It is necessary to calculate the gradient of each pixel, and then search for edge point pairs along the gradient direction or the opposite direction, and the algorithm takes a long time
②The stroke width is determined based on edge pixel pairs, depending on the character edge and the integrity of the character strokes. If the character edges are blurred or the strokes are incomplete, it is easy to cause false deletion of the candidate area.

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
  • Vehicle license plate detection and recognition method based on combination of MSER and SWT
  • Vehicle license plate detection and recognition method based on combination of MSER and SWT
  • Vehicle license plate detection and recognition method based on combination of MSER and SWT

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0060] The present invention will be described in further detail below through specific examples in conjunction with the accompanying drawings.

[0061] figure 1 is a flow chart of the method of the present invention.

[0062] 1. License plate detection

[0063] figure 2 It is a comparison picture before and after MSER segmentation; among them, (a1) is the original image, (a2) is the MSER image, (a3) ​​is the Canny image, (a4) is the Canny segmentation MSER image, (b1) is the original image, (b2) is the MSER graph, (b3) is the Canny graph, and (b4) is the Canny split MSER graph.

[0064] Step 1: MSER detection segmentation and transformation region screening.

[0065] will be like figure 2 The original color images shown in (a1) and (b1) are converted to uint8 type grayscale images (the license plate area has been intentionally blurred to protect privacy). Perform a piecewise linear grayscale transformation as follows

[0066]

[0067] Among them, (x 1 ,y 1 ) and...

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 provides a vehicle license plate detection and recognition method based on combination of MSER and SWT. TO reduce influence of a shooting angle and illumination on license plate recognition and improve the license plate recognition accuracy rate under a complicated background, the method first performs MSER extraction and Canny edge detection, and performs MSER screening after an intersection of MSER extraction and Canny edge detection is taken according to license plate character characteristics; then SWT based on morphological processing is performed in areas after screening, areas after screening are aggregated, and combined with geometrical characteristics of a license plate, fine positioning of the license plate is completed; finally, segmentation correction is performedon connected domains in a successfully positioned area, a framework is extracted and normalization is performed, and the framework after normalization is matched with a template after refining and normalization; and HU invariant moment and grid characteristics are utilized to recognize the Chinese character of the first character, and jump point scanning, statistics and encoding are adopted to recognize numbers and letters. The vehicle license plate detection and recognition method provided by the invention has relatively a high accuracy rate and robustness of license plate detection and recognition obtained under a complicated background at multiple angles and under variable illumination conditions.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition and image processing, and in particular relates to a vehicle license plate detection and recognition method based on the combination of MSER and SWT. Background technique [0002] As the number of cars continues to increase, intelligent traffic management systems are widely used, and license plate recognition is a key part of the system. The license plate recognition system mainly includes three parts: license plate location, character segmentation and character recognition. At present, the common license plate location algorithms at home and abroad are mainly based on texture features, color features, edge information, transform domain analysis and morphological processing; character segmentation algorithms mainly include projection methods, connected domain methods and prior knowledge methods; character recognition methods mainly There are template matching method, neural network me...

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): G06K9/32G06K9/34
Inventor 王艳谢广苏崔西民李宗学赵连磊
Owner UNIV OF SHANGHAI FOR SCI & 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