Illegal parking detection method based on deep learning

A parking violation and deep learning technology, applied in the field of parking violation detection based on deep learning, can solve the problems of unsatisfactory real-time effect, high hardware dependence, high deployment cost, etc., to solve the tracking timing problem, improve the calculation speed, Solve the effect of illegal parking detection

Pending Publication Date: 2020-08-14
ZHEJIANG GONGSHANG UNIVERSITY +1
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Existing illegal parking detection methods are highly dependent on hardware and require the cooperation of multiple devices. The cost of deployment is high, and the real-time effect is not ideal.

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
  • Illegal parking detection method based on deep learning
  • Illegal parking detection method based on deep learning
  • Illegal parking detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0024] A method for detecting illegal parking based on deep learning, comprising the steps of:

[0025] Step 1. Designate a no-parking zone (NPZ, no parking zone) for the surveillance video.

[0026] Step 2. Access the real-time video screen or video, detect the vehicle in the video, and obtain the vehicle detection frame, specifically:

[0027] The video is input into the yolo-mini network model to detect the vehicle in the video screen. The yolo-mini network model uses yolo-tiny as the backbone network, and the two largest calculations (Bflops) of the 12th layer and the 21st layer are used. The convolution operation between the convolution filter F of the layer and the input feature X is modified from multiplication to the L1 distance, that is, the absolute value of the difference between F and X, and a BN layer is added to normalize the result , to ensure that the original activation function can be used normally, and the modified network is named yolo-mini.

[0028] Step...

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 an illegal parking detection method based on deep learning. The method comprises the following steps: reading a video frame for detection; calculating a central point of the detected automobile detection frame; judging whether the central point is in a no-parking area or not, if so, judging whether the parking time exceeds a set value or not, and if not, reminding. According to the invention, through simplified modification of a yoto-tiny network structure, the speed of detecting a single type of targets is faster on the premise of not losing the precision; through binaryzation of the IOU, the calculation speed is significantly improved. Through a-IOU calculation, illegal parking can be more simply, conveniently and accurately determined. And by comparing the centerof the detection frame with the center stored in the set C, the tracking timing problem is simply solved.

Description

technical field [0001] The invention belongs to the field of video image processing and pattern recognition in computer vision, and relates to a method for detecting illegal parking based on deep learning. Background technique [0002] Today, the urban traffic problem has become a worldwide problem. Countries all over the world have spent a lot of manpower, material and financial resources on solving the urban traffic problem, but they cannot effectively solve this problem. Among them, the phenomenon of illegal parking is particularly serious with vehicles parked indiscriminately. A large number of vehicles occupy sidewalks and emergency passages, which bring huge harm and hidden dangers to traffic. The phenomenon of illegal parking in the city has become an important problem that needs to be solved urgently. At present, it mainly relies on camera manual monitoring and traffic police patrolling, which is not only inefficient, but also consumes too much manpower and material...

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): G06K9/00G06K9/32G08G1/017
CPCG08G1/0175G06V20/40G06V20/52G06V10/25G06V2201/08
Inventor 陶家威王慧燕陈海英
Owner ZHEJIANG GONGSHANG UNIVERSITY
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