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

Vehicle real-time overload detection method based on convolutional neural network

A technology of convolutional neural network and detection method, which is applied in the direction of neural learning method, biological neural network model, neural architecture, etc., can solve the problems of artificial inability to work 24 hours, inability to guarantee accuracy, and long time-consuming parking detection. Road non-stop overload detection, avoiding traffic jams and road traffic accidents, good real-time detection effect

Active Publication Date: 2021-06-25
迪比(重庆)智能科技研究院有限公司 +1
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This manual detection method has the following disadvantages: (1) manual work cannot be done 24 hours a day; (2) the accuracy rate cannot be guaranteed due to the subjective judgment of law enforcement personnel
(3) The efficiency is low, the parking detection takes a long time, and it is easy to cause traffic jams

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 real-time overload detection method based on convolutional neural network
  • Vehicle real-time overload detection method based on convolutional neural network
  • Vehicle real-time overload detection method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] According to the above description, the following describes the specific implementation process of the present invention.

[0042] The offline part is divided into 2 steps:

[0043] Step 1: Data Acquisition

[0044] Use the camera to collect on-the-spot and shoot from multiple angles and scenes to ensure that the photos of vehicles with each number of axles and wheelbases are covered, and the number is about 5,000.

[0045] Step 1.1: Dataset production

[0046] Label the wheels and body on the photos taken, and make them into a data set in VOC format.

[0047] Step 2: YOLO-V3 network framework construction and model training

[0048] The idea of ​​the YOLO series of algorithms is to use a picture to be detected as input, through a convolutional network, to directly classify and return the bounding box. The network structure of YOLO-V3 (as attached figure 2 Shown) consists of two parts, one is the backbone network Darknet-53 responsible for feature extraction, 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 discloses a vehicle real-time overload detection method based on a convolutional neural network. The method comprises the steps: carrying out the real-time detection of a road vehicle through employing a convolutional neural network method and employing a YOLO-V3 detection algorithm, obtaining the number of axles through the detection of the number of wheels, detecting the relative axle distance, acquiring the maximum load of the vehicle by comparison with the national standard of the vehicle load, and finally achieving real-time vehicle overload detection by comparing the actual load obtained by a piezoelectric sensor below the vehicle. According to the method, the detection real-time performance is good, the problem of road non-stop overload detection can be solved, and possible traffic jam and road traffic accidents can be avoided. A channel pruning technology is used, and a network structure is simplified under the condition that the detection precision is not influenced, so that the detection method has relatively low requirements on hardware, the equipment cost is reduced, and the method is more in line with an application scene.

Description

technical field [0001] The invention relates to target detection technology, in particular to a convolutional neural network-based detection method for trucks not stopping and overloading. Background technique [0002] The overloading of trucks in road transportation will not only affect the safety of roads and bridges, but also threaten the safety of public life. At the same time, because the load of overloaded vehicles is higher than the load of roads and bridges stipulated by the state, it will accelerate the wear and tear of roads and bridges, resulting in a large amount of maintenance funds. lead to traffic accidents. The load of overloaded vehicles generally far exceeds the design load of roads and bridges, and their frequent driving on the road will cause road surface damage, bridge breakage, and greatly shorten the normal service life of the road. [0003] Aiming at the problem of preventing vehicle overloading, my country currently mainly adopts pasting weight limi...

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): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/54G06V10/44G06V2201/08G06N3/045G06F18/23213G06F18/241G01G19/024G06N3/082G06N3/04
Inventor 王玉娟卢功林宋永端谈世磊杨雅婷任春旭刘明杨
Owner 迪比(重庆)智能科技研究院有限公司
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