China
Africa
Explore chapters and articles related to this topic
Novel safety concept at the Austrian construction lots of Brenner Base Tunnel
Published in Daniele Peila, Giulia Viggiani, Tarcisio Celestino, Tunnels and Underground Cities: Engineering and Innovation meet Archaeology, Architecture and Art, 2020
Safety planning scenarios range from classic cave-ins to fire, injuries with on-site medical treatment, gas leaks, explosion hazards, water ingress, technical malfunctions, radiation and exceedance of limit values. In principle, fires are dealt with most intensively, as only a minimal amount of time can pass before the safety-related and impact-oriented measures begin. In the other scenarios, these measures can be adapted accordingly with far fewer critical time frames. Fire seems to be the most complex yet most likely form of major accident. A wide range of safety-related measures are therefore aimed at early fire detection and reduction of the impact of fire. Specifically, the following scenarios are considered in the safety planning: Blast accidents, chemical accidents, explosions, fire smoke, gas leaks, oxygen depletion, particulate matter, low pressure, snow, avalanche, power failure, traffic and water ingress.
Novel safety concept at the Austrian construction lots of Brenner Base Tunnel
Published in Daniele Peila, Giulia Viggiani, Tarcisio Celestino, Tunnels and Underground Cities: Engineering and Innovation meet Archaeology, Architecture and Art, 2019
Safety planning scenarios range from classic cave-ins to fire, injuries with on-site medical treatment, gas leaks, explosion hazards, water ingress, technical malfunctions, radiation and exceedance of limit values. In principle, fires are dealt with most intensively, as only a minimal amount of time can pass before the safety-related and impact-oriented measures begin. In the other scenarios, these measures can be adapted accordingly with far fewer critical time frames. Fire seems to be the most complex yet most likely form of major accident. A wide range of safety-related measures are therefore aimed at early fire detection and reduction of the impact of fire. Specifically, the following scenarios are considered in the safety planning: Blast accidents, chemical accidents, explosions, fire smoke, gas leaks, oxygen depletion, particulate matter, low pressure, snow, avalanche, power failure, traffic and water ingress.
Efficient Convolutional Neural Networks for Fire Detection in Surveillance Applications
Published in Mahmoud Hassaballah, Ali Ismail Awad, Deep Learning in Computer Vision, 2020
Khan Muhammad, Salman Khan, Sung Wook Baik
Early fire detection in the context of disaster management systems during surveillance of public areas, forests, and nuclear power plants can result in the prevention of ecological, economic, and social damage. However, early detection is a challenging problem due to varying lighting conditions, shadows, and movement of fire-colored objects. Thus, there is a need for an algorithm that can achieve better accuracy in the aforementioned scenarios while minimizing the number of false alarms. To achieve this goal, we explored deep CNNs and devised a fine-tuned architecture for early fire detection during surveillance for effective disaster management. Our system is overviewed in Figure 3.2.
A New Approach to Vision-based Fire and its Intensity Computation Using SPATIO-Temporal Features
Published in IETE Journal of Research, 2023
Mirza Adnan Baig, Najeed Ahmed Khan, Muhammad Masood Rafi
Fire is one of the major and critical risks to human life and natural resources around the world. The density of urbanization is severely growing with some unwanted consequences such as overloaded living spaces. The automatic surveillance of their surroundings may help to take proactive actions to save humanity, especially in case of fire hazards and global warming that may lead to the loss of billions of dollars, such as the losses in property in china between March and April of 2019 [1]. It is important to take preventive and corrective actions in advance before happening a fire hazard [2]. Efficient fire detection and behavior anticipation system can play a pivotal role in the reduction of damages caused by fires as the devastation caused by the fire is very rapid and quite difficult to predict. These losses can be minimized if the fire detected at an early stage with a proactive approach. The last decade has witnessed the use of computer vision [3] for fire detection, early fire suppression, fire measurement. The current fire detection methods in different fire systems alarming become common and rely on various visual sensors for flame detection [4]. Fire detection on color images or in a live video is a challenging task because the images are highly affected by environmental and physical conditions [5]. Rapid development in surveillance cameras and video processing techniques has increased considerable research work on fire detection in using the computer-vision field. Although these systems can play a major role in fire detection, however, either they can be used in a limited environment or may give a false alarm. Several of the proposed algorithms for fire detection are based on color feature [6–8]. The color feature is used for fire detection generally gives more false alarm [9]. For example, a yellow color waving a flag or a yellow dominant color waving a piece of cloth may also be detected as fire.Tugaret al. [10] proposed a fire detection method using CIE L*a*b*(Commission International D’eclairage) color space obtained from RGB (Red Green Blue) color space. They have tested their approach on specific fire-related videos. The limitation of their proposed system is the assumption that the fire may spread gradually not an explosion. The sudden explosion of fire may not be identified by their proposed system.