• Fire Science and Engineering
• /
• v.28 no.2
• /
• pp.54-63
• /
• 2014

The purpose of this study is to test the adaptability of the CO sensor as a fire detector by analyzing its sensing characteristics on fire. In order to test the applicability, we designed and made a fire testing ground whose size is similar to that regulated by UL 268, carried out real fire tests suggested by UL 268, and conducted a comparison analysis on the sensing characteristics between the CO sensor and a photoelectric smoke detector by different types of fire source. The experiment result to the sensing characteristics of the CO sensor is about twice to three times faster than that of the photoelectric smoke detector, proceeding with incomplete combustion such as paper and wood fire source in the initial fire. Especially in case of wood smoldering fire, sensing characteristics of the CO sensor is very excellent.

• Journal of Sensor Science and Technology
• /
• v.33 no.3
• /
• pp.119-124
• /
• 2024

Effective fire sensing is important to protect lives and property from the disaster. In this paper, we present an intelligent visual sensing method for detecting fires based on machine learning techniques. The proposed method involves a two-step process. In the first step, fire and non-fire images are used to train a convolutional neural network (CNN), and in the next step, feature vectors consisting of 256 values obtained from the CNN are used for the learning of a support vector machine (SVM). Linear and nonlinear SVMs with different parameters are intensively tested. We found that the proposed hybrid method using an SVM with a linear kernel effectively increased the recall rate of fire image detection without compromising detection accuracy when an imbalanced dataset was used for learning. This is a major contribution of this study because recall is important, particularly in the sensing of disaster situations such as fires. In our experiments, the proposed system exhibited an accuracy of 96.9% and a recall rate of 92.9% for test image data.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1149-1151
• /
• 2003

We performed a validation study of the MODIS active fire detection algorithm using high resolution SPOT image as the reference data set. Fire with visible smoke plumes are detected in the SPOT scenes, while the hotspots in MODIS data are detected using NASA's new version 4 fire detection algorithm. The detection performance is characterized by the commission error rate (false alarms) and the omission error rate (undetected fires). In the Sumatra and Kalimantan study area, the commission rate and the omission rate are 27% and 34% respectively. False alarms are probably due to recently burnt areas with warm surfaces. False negative detection occur where there are long smoke plumes and where fires occur in densely vegetated areas.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.450-453
• /
• 2005

A forest fire occurs mainly as natural factor such as wind, temperature or human factor such as light. Recently, the most of forest fire prevention is prediction or prevision against forest fire by using remote sensing technology. However in order to forest fire prevention, the remote sensing has many limitations such as high cost and advanced technologies and so on. Therefore, we need to multisensor integration system that utilize not only remote sensing but also in-situ sensing in order to reduce large damage of forest fire though analysis of happen cause and prediction routing of occurred forest fire. In this paper we propose a multisensor integration system that offers prediction information of factors and route of forest fire by integrates collected data from remote sensor and in-situ sensor for forest fire prevention. The proposed system is based on wireless sensor network for collect observed data from various sensors. The proposed system not only offers great quality information because firstly, raw data level fuse different format of collected data from remote and in-situ sensor but also accomplish information level fusion based on result of first stage. Offered information from our system can help early prevention of factor and early prevision against occurred forest fire which transfer to SMS service or alert service into monitoring interface of administrator.

• Journal of Sensor Science and Technology
• /
• v.21 no.3
• /
• pp.211-216
• /
• 2012

Fires are the most common disaster and early fire detection is of great importance to minimize the consequent damage. Simple sensors including smoke detectors are widely used for the purpose but they are able to sense fires only at close proximity. Recently, due to the rapid advances of relevant technologies, vision-based fire sensing has attracted growing attention. In this paper, a novel visual sensing technique to automatically detect fire is presented. The proposed technique consists of multiple steps of image processing: pixel-level, block-level, and frame level. At the first step, fire flame pixel candidates are selected based on their color values in YIQ space from the image of a camera which is installed as a vision sensor at a fire scene. At the second step, the dynamic parts of flames are extracted by comparing two consecutive images. These parts are then represented in regularly divided image blocks to reduce pixel-level detection error and simplify following processing. Finally, the temporal change of the detected blocks is analyzed to confirm the spread of fire. The proposed technique was tested using real fire images and it worked quite reliably.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.747-749
• /
• 2003

For forest fire monitoring in relatively cool area like Siberia, design of Decision Support System (DSS) is proposed. The DSS is consisted of three different algorithms to detect potential fires from NOAA AVHRR image. The algorithm developed by CCRS (Canada Center for Remote Sensing) uses fixed thresholds for multi-channel information like one by ESA (European Space Agency). The algorithm of IGBP (International Geosphere Biosphere Program) involves contextual information in deriving fire pixels. CCRS and IGBP algorithms are rather liberal compared to more conservative ESA algorithm. Fire pixel information from the three algorithms is presented to the user. The user considers all these information in making decision about the location fire takes place.

• Proceedings of the KIEE Conference
• /
• 2009.04a
• /
• pp.166-168
• /
• 2009

In this paper, we studied fire sensing and position tracing system in fire region. The coordinates of fire position is communicated to fire suppression system immediately. And the fire is suppressed by this system effectively. For the purpose, the fire occurrence region images are extracted with two different CCD surveillance cameras and fire occurrence are sensed, I suggested a method to measure the fire occurrence position. It was able to be sensed early experiment result fire outbreak and, I was able to understand a fire outbreak position coordinate at the moment when a fire occurred and, I confirmed that I could accomplish warning official announcement and fire suppression in an auto.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_3
• /
• pp.1125-1137
• /
• 2020

This study aimed to propose the method of linking satellite image-based forest fire data to supplement the limitation of forest fire survey data that records only the ignition location and area of forest fire. For this purpose, a method was derived to link the fire survey data provided by the Korea Forest Service between January 2012 and December 2019 with MODIS and VIIRS image-based forest fire data. As a result, MODIS and VIIRS-based forest fire data out of 191 wildfires in the forest fire survey data were able to identify 11% and 44% of fire damage area, respectively. An average of 56% of forest damage area was extracted from VIIRS-based forest fire data compared to forest fire areas identified by high-resolution Sentinel-2A satellites. Therefore, for large-scale forest fires, VIIRS wildfire data can be used to compensate for the limitations of forest fire survey data that records only the ignition location and area.

• Fire Science and Engineering
• /
• v.27 no.2
• /
• pp.1-5
• /
• 2013

This paper studies on an arc fault interruption control circuit (AFICC) of fault voltage sensing type. The proposed voltage sensing type AFICC (VST_AFICC) is an electrical fire prevention apparatus that operates the existing circuit breaker with sensing the instantaneous voltage drop of line voltage when occurs electrical faults. The existing Earth Leakage Circuit Breaker (ELB), Molded_case Circuit Breaker (MCCB), and Residual Current Protective Devices (RCDs) used in low voltage distributing system don't have protective capability from electric arc faults to be a major factor of electrical fire. In this paper to improve such problems, a new VST_AFICC using the distortion of voltage waveform when occurs electrical faults is proposed to prevent electrical fire. There is characteristic that the control method of proposed apparatus is different from previous current sensing type. The proposed AFICC has merit that is manufactured by small size and light weight. The practicality of a new VST_AFICC is also verified through various operation analysis.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1719-1723
• /
• 2005

An automatic target sensing and triggering system for small fire arms is proposed. The system consists of an optical collector, an infrared ray sensor responsive to human body temperature, an electric actuator and a trigger mechanism. TRIZ methodologies are used to develop solutions to several contradictory problems. Experimental results on the system performance is compared with predictions.