Multi-criteria fire detection systems using a probabilistic neural network

Rose-Pehrsson, Susan L.; Shaffer, Ronald E.; Hart, Sandra G.; Williams, Frederick W.; Gottuk, Daniel T.; Strehlen, Brooke D; Hill, Scott A.

doi:10.1016/s0925-4005(00)00481-0

Cited by 64 publications

(35 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Selection of sources was based in part on previous Navy work which evaluated the development of a multi-criteria fire detection system [3][4][5][6][7]. Tables 3 and 4 respectively present the fire and nuisance sources that were used in this test series.…”

Section: Test Sourcesmentioning

confidence: 99%

An Initial Evaluation of Video-Based Fire Detection Technologies

Gottuk¹,

Harrison²,

Scheffey³

et al. 2004

View full text Add to dashboard Cite

Standard Form 298 (Rev. 8-98)Prescribed by ANSI Std. Z39.18Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Approved for public release; distribution is unlimited. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) SPONSOR / MONITOR'S ACRONYM(S) 9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) SPONSOR / MONITOR'S REPORT NUMBER(S)Computer processing and image analysis technologies have improved significantly to allow the recent development of effective video-based fire detection systems. Currently, smoke detection algorithms are the most mature. Typically, these systems are being designed and used in large facilities, outdoor locations, and tunnels. However, the technologies are also expected, with some modifications, to be effective in smaller, cluttered compartments found on ships. With the move to use onboard video surveillance, there are advantages in using the video images for other functions, such as fire detection. The video-based recognition technology also has future potential for personnel tracking, flooding detection, and physical damage assessment onboard ship as more event recognition algorithms are developed. This work represents the initial evaluation of video-based detection technologies for improved situations awareness and damage control assessment onboard Navy ships. The test results indicate that the video-based detection systems using smoke alarm algorithms can provide comparable to better fire detection than point-type smoke detectors. Damage control; Fire detections; Machine vision UnclassifiedUnclassified Unclassified UL 54

show abstract

Section: Test Sourcesmentioning

confidence: 99%

An Initial Evaluation of Video-Based Fire Detection Technologies

Gottuk¹,

Harrison²,

Scheffey³

et al. 2004

View full text Add to dashboard Cite

show abstract

“…The database for this analysis is comprised of three independent series of tests, designated as the 1) Indiana Dunes [16], 2) Navy [10,17,18], and 3) Fire Research Station (FRS) [14] data sets. Each of these data sets are described in further detail later in this section.…”

Section: Data Setsmentioning

confidence: 99%

“…The first series of tests [10,17] included numerous fire tests of a wide range of materials (smoldering and flaming) in a closed room approximately 4.1 x 6.5 x 3.6 m high (13 x 21 x 12 ft). Examples of materials used in these fire tests include flammable liquids, mattresses, electrical cable, linens, paper/cardboard, and oily rags.…”

Section: Navymentioning

confidence: 99%

Alarm Thresholds For Smoke Detector Modeling

Geiman¹,

Gottuk²

2003

Fire Saf. Sci.

View full text Add to dashboard Cite

Models have been used to estimate smoke detector alarm times by calculating smoke optical densities resulting from fires and correlating the smoke value to an alarm threshold of the detector. This work evaluated the use of smoke optical densities outside a detector as criteria for predicting smoke detector responses. Results are presented for optical density alarm thresholds corresponding to when 20, 50 and 80 percent of the detectors had alarmed in full-scale tests. The data showed that there was a large amount of variability in the measured smoke optical density values outside a detector at the time of alarm. Major variables evaluated included detector type, fire type and nominal detector sensitivity. This evaluation also examined experimental smoke optical density data at the time of detector alarms and compared these smoke values to alarm thresholds recommended in the literature.

show abstract

“…There has been a great deal of research into early and reliable detection of fires. Most of the effort has been to reduce the alarms from nuisance signals (Table 1) while responding to those signals which emanate from actual fires (Tables 2, 3) [5][6][7][8]. This selection and classification of signals is not unique, and needs to be tailored to the application.…”

mentioning

confidence: 99%

Implementing High Reliability Fire Detection in the Residential Setting

Jones

2011

Fire Technol

View full text Add to dashboard Cite

The purpose of detecting fires early is to provide an alarm prior to when there is an environment which is deemed to be a threat to people or a building. High reliability detection is based on the supposition that it is possible to utilize a small number of sensors to ascertain unequivocally that there is a growing threat either to people or to a building and provide an estimation of the seriousness of the threat. The current generation of fire detection systems is designed to respond to smoke, heat, gaseous emission or electromagnetic radiation generated during smoldering and flaming combustion. This paper will demonstrate that these same low level signals can be utilized to achieve the goal of producing earlier detection, while improving reliability. This allows us to reduce the time to detection at the same time reducing the error rate for both false alarms as well as missing fires. Large commercial fire panels are beginning to implement such advanced algorithms for reliable fast detection; however, the focus of this work is to achieve such detection in a way that can be implemented in a lost cost, low power, EPROM which will be suitable for residential systems.

show abstract

Multi-criteria fire detection systems using a probabilistic neural network

Cited by 64 publications

References 4 publications

An Initial Evaluation of Video-Based Fire Detection Technologies

An Initial Evaluation of Video-Based Fire Detection Technologies

Alarm Thresholds For Smoke Detector Modeling

Implementing High Reliability Fire Detection in the Residential Setting

Contact Info

Product

Resources

About