Real-time probabilistic classification of fire and smoke using thermal imagery for intelligent firefighting robot

Kim, Jong-Hwan; Lattimer, Brian Y.

doi:10.1016/j.firesaf.2015.02.007

Cited by 67 publications

(34 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Paugam et al [29], for example, use the hand held thermal devices to derive fire radiative power (FRP) and flame front rate of spread (ROS) in case of a wildland fire. Kim et al [30] use a probabilistic classification to classify fire and smoke regions from a single thermal camera. These results are used to autonomously navigate a firefighting robot towards the fire and are evaluated in a realistic fire set-up.…”

Section: Low-cost Video Sensing For Situational Awareness and Bim Optmentioning

confidence: 99%

See 1 more Smart Citation

Fireground location understanding by semantic linking of visual objects and building information models

Vandecasteele

Merci

Verstockt

2017

Fire Safety Journal

View full text Add to dashboard Cite

This paper presents an outline for improved localization and situational awareness in fire emergency situations based on semantic technology and computer vision techniques. The novelty of our methodology lies in the semantic linking of video object recognition results from visual and thermal cameras with Building Information Models (BIM). The current limitations and possibilities of certain building information streams in the context of fire safety or fire incident management are addressed in this paper. Furthermore, our data management tools match higher-level semantic metadata descriptors of BIM and deep-learning based visual object recognition and classification networks. Based on these matches, estimations can be generated of camera, objects and event positions in the BIM model, transforming it from a static source of information into a rich, dynamic data provider. Previous work has already investigated the possibilities to link BIM and lowcost point sensors for fireground understanding, but these approaches did not take into account the benefits of video analysis and recent developments in semantics and feature learning research. Finally, the strengths of the proposed approach compared to the state-of-the-art is its (semi-)automatic workflow, generic and modular setup and multi-modal strategy, which allows to automatically create situational awareness, to improve localization and to facilitate the overall fire understanding.

show abstract

Section: Low-cost Video Sensing For Situational Awareness and Bim Optmentioning

confidence: 99%

“…In Kim et al [30], for example, the output of such a classifier is used for autonomous firefighting robot guiding. By using Bayesian probabilistic classification on low-level image features a decent differentiating between smoke, fire and thermal reflections in infrared images could be made.…”

Section: Smoke and Fire Classificationmentioning

confidence: 99%

Fireground location understanding by semantic linking of visual objects and building information models

Vandecasteele

Merci

Verstockt

2017

Fire Safety Journal

View full text Add to dashboard Cite

show abstract

“…Paugam et al, 24 for example, use the hand held thermal devices to derive Fire Radiative Power (FRP) and flame front Rate Of Spread (ROS) in case of a wildland fire. Kim et al 25 use a probabilistic classification to identify fire and smoke regions from a single thermal camera. In combination with the proposed object detection tool (Section 3) this will contribute to the automated fireground understanding.…”

Section: Object Detection For Fire Incident Managementmentioning

confidence: 99%

Object localization in handheld thermal images for fireground understanding

et al. 2017

View full text Add to dashboard Cite

Despite the broad application of the handheld thermal imaging cameras in firefighting, its usage is mostly limited to subjective interpretation by the person carrying the device. As remedies to overcome this limitation, object localization and classification mechanisms could assist the fireground understanding and help with the automated localization, characterization and spatio-temporal (spreading) analysis of the fire. An automated understanding of thermal images can enrich the conventional knowledge-based firefighting techniques by providing the information from the data and sensing-driven approaches. In this work, transfer learning is applied on multilabeling convolutional neural network architectures for object localization and recognition in monocular visual, infrared and multispectral dynamic images. Furthermore, the possibility of analyzing fire scene images is studied and their current limitations are discussed. Finally, the understanding of the room configuration (i.e., objects location) for indoor localization in reduced visibility environments and the linking with Building Information Models (BIM) are investigated.

show abstract

“…The algorithm automatically locates fires by thermal imaging technique, where visibility is zero. This algorithm was also stablished through test data set [27]. Again, a signal traffic management system for FDS was proposed for smart cities.…”

Section: Introductionmentioning

confidence: 99%

Network-Based Real-time Integrated Fire Detection and Alarm (FDA) System with Building Automation

Anwar

Boby

Rashid

et al. 2017

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Abstract. Fire alarm systems have become increasingly an important lifesaving technology in many aspects, such as applications to detect, monitor and control any fire hazard. A large sum of money is being spent annually to install and maintain the fire alarm systems in buildings to protect property and lives from the unexpected spread of fire. Several methods are already developed and it is improving on a daily basis to reduce the cost as well as increase quality. An integrated Fire Detection and Alarm (FDA) systems with building automation was studied, to reduce cost and improve their reliability by preventing false alarm. This work proposes an improved framework for FDA system to ensure a robust intelligent network of FDA control panels in real-time. A shortest path algorithmic was chosen for series of buildings connected by fiber optic network. The framework shares information and communicates with each fire alarm panels connected in peer to peer configuration and declare the network state using network address declaration from any building connected in network. The fiber-optic connection was proposed to reduce signal noises, thus increasing large area coverage, real-time communication and long-term safety. Based on this proposed method an experimental setup was designed and a prototype system was developed to validate the performance in practice. Also, the distributed network system was proposed to connect with an optional remote monitoring terminal panel to validate proposed network performance and ensure fire survivability where the information is sequentially transmitted. The proposed FDA system is different from traditional fire alarm and detection system in terms of topology as it manages group of buildings in an optimal and efficient manner.Introduction IntroductionFire detection process for trading and housing places is a main requirement to reduce destruction of personal property due to fire incident both natural and induced. Combining the network with controller approach, an innovative improvement has been made in building automation. The real-time surveillance system not only monitors the building automation ministrations but also control the FDMS by gathering, observing and storing related information. Among the basic technologies, networking had been used for improved building automation controller construction. Research notes that the concept of networking based controller used in many building automation process are more effective in fire detection and protection [1][2][3].

show abstract

Real-time probabilistic classification of fire and smoke using thermal imagery for intelligent firefighting robot

Cited by 67 publications

References 45 publications

Fireground location understanding by semantic linking of visual objects and building information models

Fireground location understanding by semantic linking of visual objects and building information models

Object localization in handheld thermal images for fireground understanding

Network-Based Real-time Integrated Fire Detection and Alarm (FDA) System with Building Automation

Contact Info

Product

Resources

About