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

Parking violation detection method and system based on multi-feature identification

A detection method and multi-feature technology, which is applied to the traffic control system, traffic control system, instrument and other directions of road vehicles, can solve the problems of decreased control and alarm accuracy, inability to recognize license plates, and inability to be used as evidence for illegal parking penalties, saving manpower Cost, effect of improving accuracy and success rate

Active Publication Date: 2017-04-26
SHENZHEN INFINOVA
View PDF14 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] First, the existing methods and systems for capturing illegal parking rely on multiple comparisons of license plate recognition information before and after to determine whether to park illegally. Only the analysis of the vehicle license plate has certain limitations. When blocked, the license plate cannot be effectively recognized, so it cannot reflect the real attributes of the vehicle, and thus cannot be used as effective evidence for parking violations
[0004] Second, only the retrieval accuracy of license plate recognition is low, which leads to a decrease in the accuracy of deployment and control alarms, and there is a lack of efficient countermeasures for illegal parking of license plate vehicles

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
  • Parking violation detection method and system based on multi-feature identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0098] In step S3, the following steps are also included:

[0099] S301, train the Adaboost classifier according to the picture data;

[0100] S302. Obtain the features of the target in the image data according to the haar-like feature extraction algorithm;

[0101] S303. Input the features of the target into the Adaboost classifier;

[0102] S304. The Adaboost classifier classifies the features of the input target to determine whether it is a vehicle, if yes, determine the target vehicle, and enter step S4, otherwise return to step S2.

[0103] When performing vehicle detection, the vehicle detection is mainly performed by training the Adaboost classifier. Here, the haar-like feature is mainly used as the input of the classifier and the Adaboost iterative algorithm; the haar-like feature is also called the rectangular feature, and a rectangle is divided into black and white rectangular blocks. , the calculation method is to subtract the sum of all pixel values ​​​​of the wh...

Embodiment 2

[0105] In step S301, while the Adaboost classifier is being trained, it also selects target features.

[0106] The Adaboost classifier is a classifier based on the cascade classification model, which is divided into several cascades, and each cascade is composed of multiple tree classifiers. While adding feature information to the Adaboost classifier to make it learn and train, it also The target features can be screened to select the closest feature.

Embodiment 3

[0108] In step S5, the following steps are also included:

[0109] S501. Generate a center point of the target vehicle in each piece of picture data according to the acquired pieces of picture data;

[0110] According to the Adaboost classifier, each piece of picture data is detected, and the coordinate information of the vehicle target rectangle frame of each piece of picture data is obtained, and the center point coordinate of the target vehicle is obtained through the coordinate information.

[0111] S502. Calculate the displacement of the center point;

[0112] Compare the coordinates of the center point in each image with the coordinates of the center point in the next image, and obtain the displacement between the center points according to the Euclidean distance algorithm.

[0113] S503, judging whether the displacement of the center point is smaller than the threshold value, if yes, go to step S6.

[0114] If the displacement of the center point of the target vehicle...

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 parking violation detection method and system based on multi-feature identification. Tracking detection is performed on vehicles in a monitoring scope by use of an image obtaining module and a driving stop determining module, after parking target vehicles are determined, the target vehicles are amplified for capturing detail features through the image obtaining module, and features of the parking violation vehicles, including vehicle shapes, colors, parking positions and license plate numbers are identified by use of the feature identification module arranged in the system. The method and system have the following advantages: the parking violation vehicles are identified by use of the multiple features, the limitation of a conventional method of analyzing parking violation only through vehicle license plate numbers is avoided, the accuracy and the success rate of parking violation monitoring are greatly improved, and the manpower cost of traffic law enforcement is saved.

Description

technical field [0001] The invention relates to the field of intelligent video processing, in particular to a parking violation detection method and system based on multi-feature recognition. Background technique [0002] With the development of social economy, the number of motor vehicles has increased sharply, and the corresponding growth of parking spaces is relatively scarce, especially in prosperous areas and road sections, where the vehicle density is high and parking spaces are few. Parked on both sides of the road, causing traffic jams and other urban traffic problems. Due to the limited number of police officers, it is impossible to check the illegal parking behavior in the city around the clock, and illegal parking has brought great obstacles to traffic management. In order to meet the needs of traffic law enforcement, automatic detection technology and capture equipment for illegal parking of vehicles have emerged in the industry to assist the traffic police in a...

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): G08G1/017
CPCG08G1/0175
Inventor 易苗刘军臧磊
Owner SHENZHEN INFINOVA
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