Supercharge Your Innovation With Domain-Expert AI Agents!

A smoke and fire recognition algorithm based on a SSD framework

A recognition algorithm and pyrotechnic technology, applied in the field of SSD-based pyrotechnic recognition algorithm, can solve the problems of low flame detection accuracy, easy exposure of the picture, complex background, etc., to avoid over-fitting, improve alarm accuracy, and ensure accuracy.

Pending Publication Date: 2019-01-08
TIANJIN ISECURE TECH
View PDF13 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The high-speed rail surveillance video has problems such as complex background, high noise, and easy exposure of the picture. These complex environmental factors will cause the problem of low flame detection accuracy. Based on this, a firework recognition algorithm is proposed, and the technical solution adopted is as follows:

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
  • A smoke and fire recognition algorithm based on a SSD framework
  • A smoke and fire recognition algorithm based on a SSD framework
  • A smoke and fire recognition algorithm based on a SSD framework

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] like figure 1 As shown, the detection model training network in the present invention is a model network reconstructed based on the VGG16 network. The model network reduces one fully connected layer on the basis of VGG16, retains two fully connected layers fc6 and fc7, and increases the There are 6 convolutional layers conv8_1, conv8_2, conv9_1, conv9_2, conv10_1, conv10_2 and 1 pooling layer pool11.

[0023] In this embodiment, the establishment and training of the detection model are implemented based on the caffe framework. The steps of implementing the model network include:

[0024] Step 1: Create a header file, add the layer header file and place it under include / caffe / layers / . Inherited from vision_layers.hpp. Add the parameters of this layer in the caffe.proto file.

[0025] Step 2: Create the corresponding source file and place it under src / caffe / layers / , implement the LayerSetUp method, which is used to read the parameters of the layer, initialize the weig...

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 background of the high-speed railway surveillance video is complex, much noise exits, frames are liable to be exposed, so a smoke and fire recognition algorithm based on an image deep learning SSDframework is studied. A training network of a detection model is a VGG16 network, and the training network of detection model after reconstruction is additionally provided with 6 convolution layers and 1 pool layer on the basis of VGG16. The parameters to be designed for realizing the convolution layer include the number of filters, the size of the filter, the initialization method of the parameter, the initialization method of the offset, whether to turn on the offset term, the number of pixels added to each side of the input and the step size of the filter. The smoke and fire recognition system designed based on the detection model trained by the network includes a video acquisition unit, an image enhancement processing unit, a smoke and fire identification unit and a data storage unit.This system is used with high-speed railway surveillance video, and even if the image resolution is low and high-speed railway surveillance video has a complex background, the accuracy of smoke and fire object detection can be ensured.

Description

technical field [0001] The invention belongs to the field of flame detection, in particular to a fireworks identification algorithm based on SSD. Background technique [0002] In the high-speed railway carriage, once a fire alarm occurs, the loss will be immeasurable, so the prevention and timely detection of the fire alarm is the top priority. The most widely used so far are mainly temperature-sensing fire detectors and smoke-sensing fire detectors. The temperature-sensing fire detector and the smoke-sensing fire detector detect the temperature and smoke concentration around the flame, and judge whether to issue an alarm message according to the amount of the sensed information compared with the threshold value. [0003] With the development of technology, a flame and smoke recognition system based on image information has appeared. This recognition system can process the collected images in real time, which can greatly shorten the early warning time, and is conducive to 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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V20/44G06V20/41G06F18/29Y02T10/40
Inventor 张德馨史玉坤
Owner TIANJIN ISECURE TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com