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

One-stage safety helmet detection method based on S3FD network

A detection method and safety helmet technology, applied in the field of target detection, can solve problems such as difficulty in detection of safety helmets, and achieve the effect of making up for losses

Pending Publication Date: 2022-03-08
HUNAN VALIN XIANGTAN IRON & STEEL CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a one-stage safety helmet detection method based on the S3FD network to solve the problem of difficulty in safety helmet detection mentioned in the above background technology

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
  • One-stage safety helmet detection method based on S3FD network
  • One-stage safety helmet detection method based on S3FD network
  • One-stage safety helmet detection method based on S3FD network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0024] A stage of safety cap detection methods based on the S3FD network, including the following steps:

[0025] (1) First, the feature map is extracted by S3FD, forming a characteristic pyramid;

[0026] (2) Set the threshold, filter the prediction box that is less than the threshold, delete the redundancy is not conducive to the training, and the network is named filtering network;

[0027] (3) The feature of the filtering network will be sent to the connection network, so that the deep shallow characteristics are further integrated;

[0028] (4) For the characteristics of the S130 step and the sample left by S120, send a precise classification subsidiacy, implement further classification and regression, further judgment and confirming the category and location of the hard hat;

[0029] Error operation is the computational process of the loss function when convolutional neural network training.

[0030] General expressions of the loss function are:

[0031] N cls It is said th...

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

According to the one-stage safety helmet detection method based on the S3FD network, on the basis of a one-stage detector S3FD, the S3FD is changed into a network comprising a filtering structure, a connection structure and an accurate judgment structure. The network can filter redundant samples, has promotion significance for small target detection, and can improve the detection capability of the safety helmet to a great extent. In addition, a feature compensation module is provided, and the module can reduce feature loss in the convolution operation process. Therefore, the problem of few effective features in small target detection is solved, and the safety helmet detection accuracy is improved to a certain extent. Through the improvement, the training speed of the model is accelerated, and the training efficiency and the detection precision are both improved.

Description

Technical field [0001] The present invention belongs to the field of target detection, involving a phase of safety cap detection of a S3FD network. Background technique [0002] Safe Hat Detection As an important part of a smart security linkage alarm system, he has received extensive attention from people. With the continuous maturity of deep learning technology, the target testing technology is developing rapidly. Deep learning algorithm has automatic extraction characteristics, flexible structural, fast detection speed, high precision, and target detection of deep learning techniques can achieve end-to-end detection. Implementation steps can be roughly divided into two types: two-stage target detection algorithm and a phase target detection algorithm. The second-stage target detection algorithm refers to the first birth to a candidate zone, producing a pre-inspection box containing the object to be detected, and then classified by convolutional neural network. The process of t...

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
IPC IPC(8): G06V10/72G06V10/774G06V10/80G06V10/764G06V10/766G06V10/82G06V10/44G06N3/04G06N3/08G06K9/62
CPCG06N3/084G06N3/045G06F18/214G06F18/10G06F18/24G06F18/253
Inventor 佘宏彦邓明华梁理吴赟安长智廖艳曾敏曾志豪唐秋良
Owner HUNAN VALIN XIANGTAN IRON & STEEL CO LTD
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