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

Video-based tunnel traffic flow statistical method

A statistical method and technology of traffic flow, applied in the field of video-based traffic flow statistics in tunnels, can solve problems such as low accuracy and easy tracking and loss of vehicles, and achieve improved detection accuracy, high robustness, and high detection and recognition accuracy. Effect

Active Publication Date: 2021-08-31
四川九通智路科技有限公司
View PDF20 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the mean shift is tracking the target, the size of the target frame does not change with the change of the target size, which makes the vehicle easy to follow.
The target tracking algorithm based on Kalman filter believes that the motion model of the object obeys the Gaussian model, so as to predict the target motion state, and then compare it with the observation model to update the state of the moving target according to the error. The accuracy of this algorithm is not very high

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
  • Video-based tunnel traffic flow statistical method
  • Video-based tunnel traffic flow statistical method
  • Video-based tunnel traffic flow statistical method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] This embodiment discloses a video-based tunnel traffic statistics method, refer to the description attached figure 1 and figure 2 , this method is based on the camera installed in the tunnel to collect image data, and the stake number information installed by the camera identifies its location, mainly including the following steps:

[0034] A. Data set production

[0035] Obtain a number of video images containing vehicles in the tunnel, convert the collected video images into pictures using Python, and use the labelImg tool to label each picture to obtain the original picture and label data; when labeling, divide the vehicle into three categories, namely cars, trucks and buses, and finally get the original pictures and label data in xml format, and put them into the JPEGImages and Annotations folders respectively; in the process of picture annotation, delete pictures that do not contain vehicles;

[0036] B. Data set division

[0037]Divide the marked pictures in s...

Embodiment 2

[0047] This embodiment discloses a video-based tunnel traffic statistics method. The yolov3 model is mainly composed of two parts, the Darknet-53 feature extraction network and the prediction network. The Darknet-53 feature extraction network obtains a feature map. On the basis of Example 1 , in this embodiment, the feature extraction network in the yolov3 model performs feature extraction on the input picture, and extracts three feature maps of different scales 13×13, 26×26, and 52×52 for prediction (13×13 represents the picture The width and height of the picture are both 13 pixels, 26×26 means that the width and height of the picture are both 26 pixels, and 52×52 means that the width and height of the picture are both 52 pixels), in the vehicle recognition detection model of this application , the size of the picture input to the network model is 416*416*3 (representing the width of the picture is 416 pixels, the height is 416 pixels, and 3 color channels), and each scale ca...

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 relates to the technical field of vehicle detection, and discloses a video-based tunnel traffic flow statistical method, which comprises the following steps of: A, making a data set; B, dividing a data set; C, constructing a vehicle identification and detection model based on the improved yolov3; D, training a vehicle identification model; D1, performing vehicle tracking; and D2, carrying out traffic flow statistics. According to the method, the current frame of vehicle is detected by using the improved yolov3 network, the introduction of noise when the inter-frame information is used for vehicle identification is avoided, the vehicle can be accurately predicted through feature extraction according to the trained weight and offset, the method is hardly influenced by the vehicle speed, the robustness is very high, and the detection and identification accuracy of the vehicle is high.

Description

technical field [0001] The present application relates to the technical field of vehicle detection, and specifically relates to a video-based tunnel traffic flow statistics method. Background technique [0002] The current methods of traffic flow statistics are mainly based on ultrasonic detection, induction coil detection, microwave detection and video detection. Among them, ultrasonic testing equipment is relatively small and easy to install, but its performance gradually decreases with the influence of ambient temperature and airflow; induction coil testing has the advantages of standardization of product equipment and high detection accuracy, but it needs to dig out the road surface for burial during installation, which will cause problems. Blocking traffic will affect the life of the road surface; microwave detection is simple and convenient to install, will not damage the road surface, can realize all-weather detection, and has strong anti-interference ability, but has...

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): G06T7/00G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06T7/0002G06N3/08G06T2207/10016G06T2207/20076G06T2207/20081G06T2207/20084G06T2207/30232G06T2207/30236G06T2207/30242G06V20/52G06V10/44G06N3/047G06N3/045G06F18/23213G06F18/2415G06F18/241
Inventor 张蓉申莲莲邓承刚叶琳龚绍杰
Owner 四川九通智路科技有限公司
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