A fire video detection and early warning method based on image multi-feature fusion

A multi-feature fusion, video detection technology, applied in the fields of instrument, calculation, character and pattern recognition, etc., can solve the problems of insufficient detection ability of smoldering fire, inability to detect fire in time and accurately, public alarm and panic, etc., and achieve a good scope of application. and reliability, excellent generalization ability, improved extraction effect

Active Publication Date: 2022-04-22
NANJING UNIV OF POSTS & TELECOMM
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The current traditional fire detection equipment is mainly based on temperature-sensing and smoke-sensing detectors, but this kind of sensor-based detectors need to reach their detection threshold before alarming, the delay is serious, and because the detection range is small, it needs to be installed in a large space. Large-scale laying can only achieve the detection effect, resulting in high cost
In addition, in tall buildings, forests, tunnels, etc., due to the dilution of space and airflow, such equipment often cannot detect fires in time and accurately.
[0003] With the development of video surveillance systems, surveillance cameras have spread all over the streets and alleys. Combining image processing and image recognition technologies on the basis of existing video surveillance systems can not only complete the task of fire detection, but also reduce costs and improve anti-interference capabilities. It is well adapted to complex environments with large spaces and a lot of airflow; there are already many image-based fire detection methods, but these methods have the following problems: on the premise of ensuring reliable detection, the delay needs to be further improved; flame detection is the main , the ability to detect smoldering fires is insufficient; the lack of a mechanism for judging whether the detection results constitute a fire and for grading fire alarms results in panic for the public when fire is normally used or under control

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 fire video detection and early warning method based on image multi-feature fusion
  • A fire video detection and early warning method based on image multi-feature fusion
  • A fire video detection and early warning method based on image multi-feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0115] like figure 1 As shown, the present embodiment provides a fire video detection and early warning method based on image multi-feature fusion, including the following steps:

[0116] S1 performs denoising preprocessing on the acquired video image sequence using a median filter method;

[0117] S2 uses the improved method based on the gradient motion history graph to establish a background model for the preprocessed image to extract the motion area;

[0118] S3 uses the color models of flame and smoke to extract the pixels in the moving area to obtain color suspected areas;

[0119] S4 performs a morphological closing operation on the suspected color region to obtain a smooth and continuous suspected region;

[0120] S5 extracts the dynamic and static features of the suspected area;

[0121] S6 uses the extracted flame feature value as the input of the SVM classifier to judge whether it contains flame, and at the same time, calculates the extracted smoke feature and inp...

Embodiment 2

[0125] This embodiment is further optimized on the basis of Embodiment 1, specifically:

[0126] In the step S1, using the median filter for the obtained video image sequence is specifically:

[0127] Traverse and calculate the gray value of each pixel of the initial image, and calculate the median value of the gray value in the 3×3 neighborhood of each pixel to replace the original gray value of the pixel, which can effectively remove the noise in the image, Especially salt and pepper noise, which improves image quality.

[0128] In the step S2, the method based on the gradient motion history map is used to extract the motion area, and the motion area extraction result is as follows figure 2 As shown, the white is the motion area, and the black is the background, which specifically includes the following steps:

[0129] S21 first calculates the binary gradient map BGI(x, y), the calculation formula is:

[0130]

[0131] Among them, BGI x (x,y) and BGI y (x, y) are 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

The invention discloses a fire video detection and early warning method based on image multi-feature fusion. After acquiring the image sequence of the video, firstly, preprocessing is performed; then the foreground area is extracted, and then a candidate area for detection is obtained; secondly, the candidate area is Extract static features and dynamic features, and use it as the input of the SVM classifier to determine whether there is flame when detecting flames; For flame or smoke, the fire is judged according to its growth trend. When it is determined that it constitutes a fire, a fire alarm will be issued to the monitoring site, otherwise only a fire alarm will be issued in the background. The present invention can be combined with the existing monitoring system and applied to shopping malls, warehouses and other places, which reduces the cost of fire detection and early warning, has good generalization ability and applicability of the detection method, can provide reliable fire detection and early warning functions, and has practical value.

Description

technical field [0001] The invention relates to the field of image processing and recognition in computer vision, in particular to a fire video detection and early warning method based on image multi-feature fusion. Background technique [0002] Fire is one of the main threats to the safety of human life and property. With the development of society, the harm caused by fire to human society and the natural environment is becoming more and more serious. Therefore, it has become an important research topic to be able to detect fire accurately and quickly. The current traditional fire detection equipment is mainly based on temperature-sensing and smoke-sensing detectors, but this kind of sensor-based detectors need to reach their detection threshold before alarming, the delay is serious, and because the detection range is small, it needs to be installed in a large space. Large-scale laying can only achieve the detection effect, resulting in high cost. In addition, in tall bui...

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 Patents(China)
IPC IPC(8): G06V20/40G06V10/30G06V10/56G06K9/62G06V10/764G06V10/80
CPCG06V20/41G06V20/46G06V10/30G06V10/56G06F18/2411G06F18/253
Inventor 陈美娟何爱龙管铭锋
Owner NANJING UNIV OF POSTS & TELECOMM
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