Video smoke detection method based on convolutional neural network

A convolutional neural network and detection method technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as environmental interference, improve efficiency, reduce static object interference, and have a wide range of applications.

Inactive Publication Date: 2020-04-21
HARBIN UNIV OF SCI & TECH
View PDF5 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing fire smoke recognition algorithms have strong scene pertinence and are particularly susceptible to environmental interference, and propose a video smoke detection method based on convolutional neural network

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
  • Video smoke detection method based on convolutional neural network
  • Video smoke detection method based on convolutional neural network
  • Video smoke detection method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0015] A kind of video smoke detection method based on convolutional neural network of the present embodiment, described method is realized through the following steps:

[0016] Step 1, the step of preprocessing the acquired video image;

[0017] Step 2, the step of extracting the suspected smoke area from the preprocessed video image;

[0018] Step 3, the step of performing smoke feature description on the obtained suspected smoke area;

[0019] Step 4: Based on the convolutional neural network smoke texture recognition framework, the convolutional neural network method is used to perform smoke recognition on the detection area to be tested obtained in the previous step.

specific Embodiment approach 2

[0021] The difference from Embodiment 1 is that in this embodiment, a video smoke detection method based on a convolutional neural network, in the first step, the step of preprocessing the acquired video image refers to the step of preprocessing the input video Denoise the image, select the appropriate color space, extract key frames, etc., to improve the anti-interference ability of the target area, specifically:

[0022] First, the video image of the smoke scene is acquired by the camera;

[0023] Then, the collected sequence images are processed by the background subtraction method, and the foreground image of the moving target is preliminarily extracted;

[0024] Finally, noise interference in the foreground image is removed by morphology.

specific Embodiment approach 3

[0026] The difference from the second specific embodiment is that in this embodiment, a video smoke detection method based on a convolutional neural network, in the second step, the step of extracting the suspected smoke area from the preprocessed video image refers to, Analyze the characteristics of smoke movement, including: flame color, flame gray scale, delay area, delay color, smoke shape and smoke radiation intensity, so as to segment the area to be detected and reduce the amount of calculation.

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 video smoke detection method based on a convolutional neural network, and belongs to the field of picture recognition. Existing fire smoke identification algorithms have strong scene pertinence and are particularly susceptible to environmental interference. The invention discloses a video smoke detection method based on a convolutional neural network. The method comprisesthe following steps: preprocessing an acquired video image; performing suspected smoke area extraction on the preprocessed video image; carrying out smoke feature description on the obtained suspected smoke area; and based on a convolutional neural network smoke texture recognition framework, performing smoke recognition on the to-be-detected area obtained in the previous step by utilizing a convolutional neural network method. The moving target in the foreground image is input into the CNN model for smoke identification, so that the smoke detection efficiency is improved while the static object interference is reduced.

Description

technical field [0001] The invention relates to a video smoke detection method based on a convolutional neural network. Background technique [0002] The research on fire detection can be traced back more than 30 years ago. The earliest scholars used flame as the research object, and the color and brightness of the flame were the main features, so the research on color preprocessing appeared, from RGB color model, HSV color model Model, and YCbCr, but the detection effect of this kind of method is not accurate, and the early warning is often only after the wild fire occurs in a wide range, and then gradually began to explore the smoke accompanying the fire. For smoke features, smoke detection in domestic video images started late, and gradually formed a research situation with different emphases. Yaqin Zhao, Yuan Feiniu and others used LBP, LBPV, and color space to extract local features for recognition. Li Sun et al. used dark channel priors to enhance color through a cer...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/52G06N3/045G06F18/241
Inventor 贺雅楠于洪丹
Owner HARBIN UNIV OF SCI & TECH
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