Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium

A detection method and helmet technology, which is applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve problems such as large amount of calculation, high requirements for shooting angle and image quality, redundant frames, etc., and achieve detection Low requirements and the effect of reducing the amount of network parameters

Active Publication Date: 2021-06-15
NORTHWEST UNIV(CN)
View PDF6 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In exploring the helmet detection method based on deep learning, Prajwal et al. first used YOLOv2 to detect people, motorcycles and electric vehicles in the video frame, and then used YOLOv3 to detect whether there is a helmet in the Region of interest (ROI). The framework used by this method is redundant and the amount of calculation is large; Noel et al. first use the traditional machine vision method to classify motorcycles and electric vehicles, and then use the YOLOv3 target detection framework to detect helmets. This method has requirements for shooting angles and image quality. higher

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
  • Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium
  • Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium
  • Helmet detection method and device based on improved YOLOv5s, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] See figure 1 , figure 1 A flowchart of a helmet detection method based on improved YOLOv5s provided by the embodiment of the present invention. This embodiment discloses a helmet detection method based on improved YOLOv5s, including:

[0062] Step 1. Build the YOLOv5s-Light target detection framework.

[0063] Specifically, this embodiment improves the algorithm of the first YOLOv5s network, adopts depth separable convolution module, SE (Squeeze-and-excite) attention module and improved inversion residual in network model design, and activates The function adopts the H-swish function. While ensuring the detection frame rate and accuracy, a lightweight network model YOLOv5s-Light target detection framework was obtained, which greatly reduced the number of model parameters, improved the efficiency of the model, and realized a lightweight and easy-to-deploy helmet Detection method.

[0064] Further, step 1 includes:

[0065] Step 1.1. Based on the lightweight network...

Embodiment 2

[0127] See Image 6 , Image 6 A schematic structural diagram of a helmet detection device based on the improved YOLOv5s provided by the embodiment of the present invention.

[0128] This embodiment discloses a helmet detection device based on improved YOLOv5s, including:

[0129] Model building block 1, used to build the YOLOv5s-Light target detection framework;

[0130] Model training module 2, for utilizing the target image training set to train the YOLOv5s-Light target detection framework to obtain the YOLOv5s-Light target detection training framework;

[0131] The information processing module 3 is used to input the image set to be detected into the YOLOv5s-Light target detection training framework to obtain the target detection set;

[0132] The riding detection module 4 is used to detect the target detection set by using the riding detection algorithm to obtain the target detection result.

[0133] In one embodiment of the present invention, model construction modul...

Embodiment 3

[0138] See Figure 7 , Figure 7 A schematic structural diagram of a helmet detection electronic device based on the improved YOLOv5s provided by the embodiment of the present invention.

[0139] This embodiment discloses a helmet detection electronic device based on improved YOLOv5s, comprising: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus;

[0140] memory for storing computer programs;

[0141] The processor is configured to implement the method steps of any one of the present embodiments when executing the computer program.

[0142] An electronic helmet detection device based on the improved YOLOv5s provided by the embodiment of the present invention can execute the above-mentioned method embodiment, and its implementation principle and technical effect are similar, and will not be repeated here.

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 a helmet detection method and device based on improved YOLOv5s, electronic equipment and a storage medium. The method comprises the following steps: constructing a YOLOv5s-Light target detection framework; using a target image training set to train the YOLOv5s-Light target detection framework to obtain a YOLOv5s-Light target detection training framework; inputting an image set to be detected into the YOLOv5s-Light target detection training framework to obtain a target detection set; and detecting the target detection set by using a riding detection algorithm to obtain a target detection result. According to the method, firstly, a YOLOv5s-Light target detection framework is constructed, the YOLOv5s-Light target detection framework greatly reduces the network parameter quantity on the basis of the YOLOv5s target detection framework, the construction of a lightweight model is realized, then the YOLOv5s-Light target detection framework is utilized to obtain a target detection set, and finally, the target detection set is detected through a riding detection algorithm to obtain a target detection result, so that a required target can be effectively detected, unnecessary targets can be filtered, and the detection requirement is low.

Description

technical field [0001] The invention belongs to the technical field of machine vision applications, and in particular relates to a helmet detection method, device, electronic equipment and storage medium based on the improved YOLOv5s. Background technique [0002] In recent years, motorcycles and electric vehicles are important means of transportation for people's daily travel, and the traffic accidents caused by them are also increasing. The death rate caused by head trauma after traffic accidents among motorcycle and electric vehicle riders is very high, about 80% are craniocerebral injuries, and the important reason for this situation is that the rider fails to wear the helmet correctly and neglects the protection of the head. [0003] Scientific research shows that if a safety helmet is worn correctly, when an accident occurs, the helmet can absorb most of the impact energy and reduce the head injury caused by the accident, thereby reducing the risk of death in traffic 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): G06N3/04G06N3/08
CPCG06N3/08G06V2201/07G06N3/045Y02T10/40
Inventor 汪霖曹世闯陈莉宜超杰张万绪
Owner NORTHWEST UNIV(CN)
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