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

Smoke detection method of integrated convolutional neural network

A convolutional neural network and detection method technology, applied in the field of smoke identification, can solve the problems of high smoke monitoring cost, insufficient accuracy and low efficiency, and achieve the effects of reducing smoke detection cost, improving accuracy and saving capital.

Pending Publication Date: 2020-04-03
WUHAN TEXTILE UNIV
View PDF16 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Solve the problems of high cost, untimely, insufficient accuracy and low efficiency of smoke monitoring

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
  • Smoke detection method of integrated convolutional neural network
  • Smoke detection method of integrated convolutional neural network
  • Smoke detection method of integrated convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Such as figure 1 , 4 As shown in ~5, a smoke detection method integrating convolutional neural network, including a suspected smoke acquisition module, a suspected smoke confirmation module and a smoke alarm module, the specific implementation steps are:

[0045] S1. The suspected smoke acquisition module acquires images containing suspected smoke after the real-time acquired images are detected by the Faster R-CNN model;

[0046] S2. The suspected smoke detection module uses the convolutional neural network to detect the candidate area of ​​the suspected smoke image, determines the image containing smoke, and uses the convolutional neural network (CNN) model to determine the size and position of the candidate area according to the center of the candidate area as a new candidate The center of the candidate area is regenerated, and then the new candidate area is identified by the convolutional neural network. Figure 4 Smoke recognition was performed on the Faster R-CN...

Embodiment 2

[0049]On the basis of the original Faster R-CNN model, the network structure is simplified, and the original 13-layer convolution and 13-layer ReLU functions are reduced to 10-layer convolution and 10-layer ReLU functions. After the reduction, the effect of feature extraction remains unchanged. , but the amount of calculation becomes smaller, which reduces the overall false positives and improves the accuracy.

[0050] After being processed by Faster R-CNN, the suspected smoke area is obtained, and the area selected by the frame on the image is the suspected smoke area. Then, the convolutional neural network is used to determine whether there is smoke in the suspected smoke image, and finally the result is processed.

Embodiment 3

[0052] Such as figure 2 As shown, the specific implementation steps of the suspected smoke acquisition module in step S1 to obtain images containing suspected smoke after passing the Faster RCNN model detection are as follows:

[0053] S11. Obtain the smoke detection Faster R-CNN model after training through the Faster R-CNN network according to a large number of smoke data samples;

[0054] S12. Use the trained smoke detection Faster R-CNN model to detect real-time video target images, and obtain images of suspected smoke candidate areas;

[0055] S13. According to the step S12, obtain the image containing the suspected smoke candidate frame, and according to the smoke similarity score corresponding to each image, when the score is greater than a certain threshold, it is determined to be smoke.

[0056] The Faster R-CNN network structure is composed of three neural networks including CNN (Convolutional Neural Networks, convolutional neural network), RPN (Region Proposal Net...

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 smoke detection method of an integrated convolutional neural network. The system comprises a suspected smoke acquisition module, a suspected smoke confirmation module and a smoke alarm module. The suspected smoke acquisition module detects an image acquired in real time through a Faster R-CNN model, and acquires an image containing suspected smoke; the suspected smoke detection module detects a candidate area of the suspected smoke image by using a convolutional neural network; the smoke alarm module responds to the processing result of the suspected smoke confirmationmodule, so the smoke detection cost is greatly reduced, the recognition efficiency and accuracy are improved, the recognition flexibility is high, no extra storage and calculation expenditure exists,the complexity of the system is reduced, and the system is efficient and energy-saving.

Description

technical field [0001] The invention relates to the field of smoke recognition by computer, in particular to a smoke detection method integrating a convolutional neural network. Background technique [0002] The open-air burning of straw belongs to low-temperature incineration and incomplete combustion. The flue gas contains a large amount of carbon monoxide, carbon dioxide, nitrogen oxides, photochemical oxidants and suspended particles, which cause air pollution and will aggravate the occurrence of smog to a certain extent. Not only that, fires caused by straw burning in various regions occur from time to time in the harvest season. Every summer and autumn harvest season, the Ministry of Environmental Protection, the Ministry of Agriculture and local governments will invest a lot of manpower to monitor the burning of straw. Satellites are used to monitor straw burning, and the monitored fire points are summarized and released. However, the phenomenon of straw burning is st...

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): G06T7/00G06N3/04G08B17/10
CPCG06T7/0002G08B17/10G06T2207/10004G06T2207/20081G06T2207/30108G06N3/045
Inventor 余锋姜明华周昌龙马佩宋坤芳
Owner WUHAN TEXTILE UNIV
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