Search results

Fire smoke detection in petrochemical plant can prevent fire and ensure production safety and life safety. The purpose of this paper is to solve the problem of missed detection and false detection in flame smoke detection under complex factory background.

This paper presents a flame smoke detection algorithm based on YOLOv5. The target regression loss function (CIoU) is used to improve the missed detection and false detection in target detection and improve the model detection performance. The improved activation function avoids gradient disappearance to maintain high real-time performance of the algorithm. Data enhancement technology is used to enhance the ability of the network to extract features and improve the accuracy of the model for small target detection.

Based on the actual situation of flame smoke, the loss function and activation function of YOLOv5 model are improved. Based on the improved YOLOv5 model, a flame smoke detection algorithm with generalization performance is established. The improved model is compared with SSD and YOLOv4-tiny. The accuracy of the improved YOLOv5 model can reach 99.5%, which achieves a more accurate detection effect on flame smoke. The improved network model is superior to the existing methods in running time and accuracy.

Aiming at the actual particularity of flame smoke detection, an improved flame smoke detection network model based on YOLOv5 is established. The purpose of optimizing the model is achieved by improving the loss function, and the activation function with stronger nonlinear ability is combined to avoid over-fitting of the network. This method is helpful to improve the problems of missed detection and false detection in flame smoke detection and can be further extended to pedestrian target detection and vehicle running recognition.

The purpose of this paper is to develop an early fire-alarm raising system based on video processing, and combine it with the omnidirectional projecting system. It not only gives alarm immediately in early fire so that people can be able to strive for more time to escape from the spot, but also solves problem of discontinued screen which was presented fire scene.

The smoke detection system is made by image processing. The flowchart of smoke detection is improved, which the method of background updating can filter out the moving objects that only stay for a short time in the image; and avoids these objects being determined to be the background. Moreover, the authors extract the flickering area to separate the non-smoke and smoke from the candidate of smoke regions. Finally, the image processing is applied in omnidirectional projecting system, then presented the 360-degree fire scene.

The results show that the smoke detection system can accurately detect the smoke and mark its location, then combining it with the omnidirectional projecting system, although the resolution of omnidirectional projecting system is not enough, it can present a continued screen and location of smoke on the 360-degree cylindrical screen.

This paper develops the smoke detection based on a improved method of image processing, and the control center staff can see the 360-degree fire scene via omnidirectional projecting system, so shorten the time to find the source of smoke.

Fire is a common disaster. Even though simple sensors such as those detecting smoke or heat are popularly employed, they require close proximity to fire. In order to obtain more reliable and more complete information, fire detection by vision sensing has recently acquired increasing attention. In the vision-based fire sensing, colour is usually used as an important cue for flame detection. However, considering there are still a large number of black-and-white (B/W) CCTV cameras installed for security purposes, a technique that can detect flame reliably in grey-scale images will be useful to protect human lives and property from the fire disaster. The paper aims to discuss these issues.

This article describes the automatic detection of fire flames in the grey-scale image sequences by a two-level image processing scheme: pixel-level and frame-level. In pixel-level processing, an evaluation function is devised to extract pixels that possibly belong to the flame region, particularly to its boundaries. Extracted fire pixel candidates are verified in frame-level processing by monitoring their distribution variations in sequential images. A circle is fitted to the candidate pixels in each image for efficient monitoring, and the presence of flame is reasoned when the position and size of the circle increase with high fluctuations.

Experimental results show that the proposed method can detect flame quite reliably using the intensity information and its temporal variations in grey-scale image sequences.

This paper presents a novel technique of vision-based flame detection. Unlike most existing techniques, the proposed technique is based on the grey-scale images of a B/W camera. To the best of the author's knowledge, it may be the first of its kind developed for general application to indoor and outdoor scenes.

At a time of soaring labour costs, price controls, dividend restraint and a host of other economic problems, fire prevention and control is an issue which management might be tempted to push down the priority list. Yet a glance at the staggering increase in fire damage costs provides a forceful reminder that this is a hazard which companies ignore at their peril. Apart from the human dangers, a full‐scale blaze can achieve in a matter of hours what adverse trading conditions may have failed to do over months or years — put you out of business. Here, in a comprehensive guide to fire prevention and control, specialists report on aspects ranging from insurance to equipment. But the underlying theme is the same: that with a little more thought and care, multi‐million pound investments could literally be prevented from going up in smoke.

The purpose of this paper is to investigate a set of preventive measures required for mitigating fire risks in big box retail facilities.

The paper identifies the potential sources of ignition and fuel in big box retail facilities. It describes the variety of hazardous situations commonly found in such facilities worldwide. The paper then endeavors to discuss the series of fire protection challenges that could be faced during fire emergencies. It also explores the challenges of evacuation and rescue in such mega store facilities.

Mega stores, distribution centers and large retail stores are amongst the most challenging occupancies from a fire protection perspective. Fires can occur in big box retail facilities at any time and from a number of causes. These facilities represent a type of occupancy that poses considerable challenges to both fixed fire suppression systems and fire departments in cases of fire emergencies. The paper also describes the responsibilities of building management staff towards their employees and the public. Facility managers should always seek proactive measures to reduce the risk of fire and fire spread in big box retail facilities. These measure include providing sufficient number and capacity of exits; clear exit access; efficient smoke detection systems; voice communication systems; and efficient automatic sprinkler systems.

This paper serves to increase the awareness about fire and its effects in mega store facilities. The paper provides practical value to property directors and facility managers responsible for the daily operations of mega store facilities and for surveyors inspecting such properties.

This paper aims to determine the halon concentration time-evolution inside an aircraft cargo compartment to design fire extinguishing systems.

A fire suppression system is numerically simulated using the lumped parameter approach.

The halon volumetric concentration, halon and air mass fluxes and the cargo compartment pressure are numerically calculated. It also determines the time to halon concentration to achieve the fire suppressant value (high pressure bottle) as well as its inerting volumetric concentration (low pressure bottle).

In the lumped parameter approach, the dependent variables of interest are a function of time alone, and its spatial distribution is neglected.

This study predicts the fire extinguishing agent behavior aiming to satisfy cargo compartment certification requirements.

This paper uses a simplified methodology, but it represents a very useful tool during the preliminary stages of the aircraft fire suppression systems design.

This paper aims to describe the sensors used in fire detection products and systems.

Following an introduction, this paper discusses the sensors used for fire detection according to the variables measured. It considers the sensing of smoke, temperature, flames and combustion products and covers residential and industrial uses, together with a short discussion of wildfire detection methods.

This shows that a multitude of different sensing techniques are used to detect fires and no single method is universally applicable. This reflects both the differing types of fires that can occur and also the many applications and environments in which fire detectors and systems are deployed.

This paper provides details of the sensor technologies and techniques used to detect fires in residential, industrial and natural environments.

This paper aims to discuss how acrylic resin influences the smoke generation of intumescent flame retardant coatings.

Thermal decomposition kinetics is used in this study to simulate the burning process. The thermal decomposition of acrylic resin can be identified in the intumescent coatings through the multi-peak fitting of derivative thermogravimetric (DTG) curves. The dormant influence of acrylic resin, combined with the smoke density, is calculated.

Multiple peaks fitting method of DTG curves helps estimate the decomposition process of acrylic resin in flame retardant coating. Combining DTG data with the smoking curve, smoking generation of acrylic resin during the combustion could be evaluated. The decomposition conversion rate of acrylic resin is 21.13 per cent. Acrylic resin generates 34.64 per cent of the total amount of smoke produced during the combustion of intumescent flame retardant coatings.

All the other intumescent flame retardant coating systems could be studied using the same approach as that used in this work to achieve an improved understanding of the smoke generation process during combustion.

The method developed here provided a simple and practical solution to analyse the decomposition and smoking generation of acrylic resin in the coating mixtures. It also can be used to analyse any thermal decomposition process of any mixed compounds.

The analysis method to evaluate resin’s smoking generation of coating’s total generation is novel, and it could be applied in all kinds of coatings and mixtures to estimate the smoking generation of one composition.

Purpose – The purpose of this paper is to develop prototype of the web-based home fire early warning system using Wiznet W5500 Ethernet module. This system protocol helps users in sending information of fire through the internet with the internet of things (IoT) method using Wiznet Ethernet module as communication media to the user.

Design/Methodology/Approach – This paper presents the design of web-based home fire early warning system using Wiznet W5500 Ethernet module. The system prototype is built using flame sensors, MQ-02 smoke sensors, and LM35 temperature sensors as input components. While on the processor side using Arduino Uno microcontroller as sensor data processing. Processed data is sent to the Ethernet module as a web server resulting in a web-based early warning information system with an alarm notification on the browser along with home location status information and sensor data.

Findings – This research produces a prototype of the web-based home fire early warning system using Wiznet W5500 Ethernet module that has been able to provide notification to the security officer housing.

Research Limitations/Implications – In the implementation of measurement, the information system only accesses one house detector or one fire location.

Practical Implications – This research produces a prototype of the web-based home fire early warning system using Wiznet W5500 Ethernet module that has been able to distribute data of temperature, smoke, and fire.

Originality/Value – The development of fire monitoring systems using flame sensors, smoke sensors and integrated temperature sensors in internet-based communication systems of things via the Internet W5500 does not appear to have been published yet.

The purpose of this paper is to present a deep ensemble neural network model for the detection of forest fires in aerial vehicle videos.

Presented deep ensemble models include four convolutional neural networks (CNNs): a faster region-based CNN (Faster R-CNN), a simple one-stage object detector (RetinaNet) and two different versions of the you only look once (Yolo) models. The presented method generates its output by fusing the outputs of these different deep learning (DL) models.

The presented fusing approach significantly improves the detection accuracy of fire incidents in the input data.

The computational complexity of the proposed method which is based on combining four different DL models is relatively higher than that of using each of these models individually. On the other hand, however, the performance of the proposed approach is considerably higher than that of any of the four DL models.

The simulation results show that using an ensemble model is quite useful for the precise detection of forest fires in real time through aerial vehicle videos or images.

By this method, forest fires can be detected more efficiently and precisely. Because forests are crucial breathing resources of the earth and a shelter for many living creatures, the social impact of the method can be considered to be very high.

This study fuses the outputs of different DL models into an ensemble model. Hence, the ensemble model provides more potent and beneficial results than any of the single models.