A Deep Separable Convolutional Neural Network for Multiscale Image-Based Smoke Detection

2022 ◽  
Author(s):  
Yinuo Huo ◽  
Qixing Zhang ◽  
Yang Jia ◽  
Dongcai Liu ◽  
Jinfu Guan ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Smoke Detection

Convolutional neural network for video fire and smoke detection

2016 ◽  
Author(s):  
Sebastien Frizzi ◽  
Rabeb Kaabi ◽  
Moez Bouchouicha ◽  
Jean-Marc Ginoux ◽  
Eric Moreau ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Smoke Detection

scholarly journals Detecção de fogo e fumaça em vídeos usando remoção de fundo e Redes Neurais Convolucionais

2020 ◽  
Author(s):  
Mayla Toshimi Nagai ◽  
Bruno M N Souza
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Smoke Detection ◽  
Fire Smoke

Em ambientes abertos, quando há baixa umidade relativa do ar e calor excessivo uma pequena faísca pode ser o gatilho para que incêndios de grandes proporções aconteçam. Diante da imprevisibilidade do surgimento do fogo, detectá-lo de forma precoce é de suma importância para agilizar o combate ao incêndio e então minimizar suas consequências. As pesquisas com as Redes Neurais Convolucionais - CNN (Convolutional Neural Network) - vem aumentando de forma exponencial com utilizações de diversas tecnologias para a detecção de fogo. Desta maneira, o presente trabalho apresenta um processo de detecção de fogo e fumaça em vídeos baseado na utilização de métodos de remoção de fundo e redes neurais convolucionais para a detecção. Como resultados preliminares, o processo proposto foi capaz de alcançar uma acurácia de 92,73% na arquitetura FSDN-Fire Smoke Detection Network e 94,88% utilizando a arquitetura XCeption. Resultados similares aos encontrados na literatura utilizando a mesma base de Vídeos mostrando que aliar a utilização de CNN após a remoção de fundo em vídeos mostra-se uma estratégia promissora para a detecção de fogo e fumaça.

scholarly journals A Robust and Dynamic Fire Detection Algorithm using Convolutional Neural Network

2021 ◽  
pp. 6-10
Author(s):  
Mrs. K. Sivasankari ◽  
◽  
Shubham Singh ◽  
Kanhaiya Kumar ◽  
Aman Dubey ◽  
...  
Keyword(s):  
Neural Network ◽  
Image Processing ◽  
Convolutional Neural Network ◽  
Major Part ◽  
Fire Detection ◽  
Data Preprocessing ◽  
Detection Algorithm ◽  
Direction Vector ◽  
Smoke Detection ◽  
Major Process

The major part of the underlying idea is going to detect the fire from upcoming smoke and the shade color of the smoke using convolutional neural network. The fire detection followed by the smoke detection is going to depend on the shade and the direction vector analysis in this paper. Image processing from the available set of data is very vague ideation so in order to strengthen the idea we are incorporating two main features that is the smoke shade and direction vector. For this major process we will involve data preprocessing through bi-variate hypothesis to select two variables as the color of smoke and the direction of the smoke and hence do the further analysis on other features that how are they going to help in the upcoming detection neurons for the robust algorithm of fire detection.

scholarly journals A Study on Flame and Smoke Detection Algorithm Using Convolutional Neural Network Based on Deep Learning

2020 ◽  
Vol 20 (1) ◽  
pp. 223-232 ◽  
Author(s):  
Jinkyu Ryu ◽  
Dongkurl Kwak
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Large Scale ◽  
Region Of Interest ◽  
Detection Algorithm ◽  
Smoke Detection ◽  
Conventional Methods ◽  
Learning Data

Recently, cases of large-scale fires, such as those at Jecheon Sports Center in 2017 and Miryang Sejong Hospital in 2018, have been increasing. We require more advanced techniques than the existing approaches to better detect fires and avoid these situations. In this study, a procedure for the detection of fire in a region of interest in an image is presented using image pre-processing and the application of a convolutional neural network based on deep-learning. Data training based on the haze dataset is included in the process so that the generation of indoor haze smoke, which is difficult to recognize using conventional methods, is also detected along with flames and smoke. The results indicated that fires in images can be identified with an accuracy of 92.3% and a precision of 93.5%.

scholarly journals A real-time video smoke detection algorithm based on Kalman filter and CNN

2021 ◽  
Author(s):  
Alessio Gagliardi ◽  
Francesco de Gioia ◽  
Sergio Saponara
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Large Scale ◽  
Moving Objects ◽  
Time Window ◽  
Hybrid Approach ◽  
Detection Algorithm ◽  
Regions Of Interest ◽  
Raspberry Pi ◽  
Smoke Detection

AbstractSmoke detection represents a critical task for avoiding large scale fire disaster in industrial environment and cities. Including intelligent video-based techniques in existing camera infrastructure enables faster response time if compared to traditional analog smoke detectors. In this work presents a hybrid approach to assess the rapid and precise identification of smoke in a video sequence. The algorithm combines a traditional feature detector based on Kalman filtering and motion detection, and a lightweight shallow convolutional neural network. This technique allows the automatic selection of specific regions of interest within the image by the generation of bounding boxes for gray colored moving objects. In the final step the convolutional neural network verifies the actual presence of smoke in the proposed regions of interest. The algorithm provides also an alarm generator that can trigger an alarm signal if the smoke is persistent in a time window of 3 s. The proposed technique has been compared to the state of the art methods available in literature by using several videos of public and non-public dataset showing an improvement in the metrics. Finally, we developed a portable solution for embedded systems and evaluated its performance for the Raspberry Pi 3 and the Nvidia Jetson Nano.

scholarly journals Automatic Smoke Detection Based on SLIC-DBSCAN Enhanced Convolutional Neural Network

IEEE Access ◽  
2021 ◽  
Vol 9 ◽  
pp. 63933-63942
Author(s):  
Dali Sheng ◽  
Jinlian Deng ◽  
Jiawei Xiang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Smoke Detection
Sign in / Sign up

Export Citation Format

Share Document