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

Lightweight safety helmet detection method and system for mobile terminal

A mobile terminal and detection method technology, applied in the field of helmet detection, can solve the problems of model loss calculation and convergence, complex network structure, large amount of parameters, etc., and achieve the effect of reducing the amount of model parameters, simple network structure, and outstanding superiority.

Pending Publication Date: 2022-02-18
QILU UNIV OF TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] 1. Due to the complex network structure and large amount of parameters, the weight file of the network model is large and occupies a large memory space. It is difficult to deploy applications on some mobile devices, such as smart phones, tablet computers and smart cameras;
[0005] 2. The existing network model is deployed to the mobile terminal to realize the helmet detection. Because the network model has the problem of model loss calculation and convergence difficulty, the loss of model calculation accuracy and speed is 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
  • Lightweight safety helmet detection method and system for mobile terminal
  • Lightweight safety helmet detection method and system for mobile terminal
  • Lightweight safety helmet detection method and system for mobile terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0060] This embodiment provides a lightweight helmet detection method for a mobile terminal, including:

[0061] Obtain relevant image information;

[0062] According to the obtained relevant image information and the preset helmet detection model, the detection result of the helmet is obtained;

[0063] Among them, the hard hat detection model is obtained by improving the residual network of Darknet53 in the YOLOv3 network, specifically: according to the CspNet network architecture, the number of channels is divided into two parts, the first part and the second part, and the first part does not make any rolls. Product operation; the second part performs two convolution and Concat operations, and superimposes the feature map after the convolution operation with the second part, and concats the superimposed feature map with the first part.

[0064] Specifically, in this embodiment, the helmet detection includes the following steps:

[0065] S1: Make model training data sets a...

Embodiment 2

[0124] In this embodiment, the experimental rendering based on the pytorch framework includes the following steps:

[0125] In this embodiment, image twisting and random rotation operations are performed on the data set used for feature enhancement; the data set is a data set of 7950 hard hats, and the data set comes from network downloads, real shots on the construction site, and open source data marked by others Set, the dataset is marked as two categories, when a helmet is detected, it is displayed as hat; when no helmet is detected, it is displayed as hanger. At the same time, the data set is converted into a data format that yolo can easily handle. The main operation is: for the open source data set downloaded from the network, convert the txt label format into XML format; Save tag types in XML format.

[0126] After preprocessing the data set, the virtual environment is built. The present invention mainly builds a virtual environment through anaconda and names it torch,...

Embodiment 3

[0131] This embodiment provides a lightweight helmet detection system for a mobile terminal, including an acquisition module and a detection module;

[0132] The acquisition module is configured to: acquire relevant image information;

[0133] The detection module is configured to: obtain the detection result of the helmet according to the obtained related image information and the preset helmet detection model;

[0134] Among them, the hard hat detection model is obtained by improving the residual network of Darknet53 in the YOLOv3 network, specifically: according to the CspNet network architecture, the number of channels is divided into two parts, the first part and the second part, and the first part does not make any rolls. Product operation; the second part performs two convolution and Concat operations, and superimposes the feature map after the convolution operation with the second part, and concats the superimposed feature map with the first part.

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 lightweight safety helmet detection method and system for a mobile terminal. The method comprises steps of obtaining related image information; obtaining a detection result of the safety helmet according to the obtained related image information and a preset safety helmet detection model, wherein the safety helmet detection model is obtained by improving a Darknet53 residual network in a YOLOv3 network, specifically, the number of channels is divided into a first part and a second part according to a CspNet network architecture, and the first part is not subjected to any convolution operation; the second part is subjected to convolution and Concat operation twice, the feature map after convolution operation and the second part are superposed, and the superposed feature map and the first part are subjected to concat connection; according to the lightweight target detection algorithm provided by the invention, the network model structure is improved, the purposes of simple network structure and small parameter quantity are realized, the influence on the time delay problem is relatively small, and the lightweight target detection algorithm is suitable for mobile equipment or scenes which cannot be applied by the original algorithm.

Description

technical field [0001] The disclosure belongs to the technical field of helmet detection, and in particular relates to a lightweight helmet detection method and system for a mobile terminal. Background technique [0002] With the advancement of science and technology, the development of deep learning has made it possible to implement many computer vision tasks. These tasks have penetrated into all walks of life, such as industrial safety, including tasks such as helmet wearing detection, falling object detection and abnormal accident detection (pedestrians). Can't afford to fall, etc.) and so on. [0003] The inventors of the present disclosure found that the network model also has the following problems in the application of helmet detection: [0004] 1. Due to the complex network structure and large amount of parameters, the weight file of the network model is large and occupies a large memory space. It is difficult to deploy applications on some mobile devices, such as s...

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): G06V20/10G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/214
Inventor 邓立霞李洪泉刘海英李铁牛
Owner QILU UNIV OF TECH
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