scholarly journals An Improvement of the Fire Detection and Classification Method Using YOLOv3 for Surveillance Systems

Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6519
Author(s):  
Akmalbek Abdusalomov ◽  
Nodirbek Baratov ◽  
Alpamis Kutlimuratov ◽  
Taeg Keun Whangbo
Keyword(s):  
Real Time ◽  
High Speed ◽  
Fire Detection ◽  
Classification Performance ◽  
Surveillance Systems ◽  
Detection Systems ◽  
Fire Detector ◽  
Board Level ◽  
In Fire ◽  
Shape And Size

Currently, sensor-based systems for fire detection are widely used worldwide. Further research has shown that camera-based fire detection systems achieve much better results than sensor-based methods. In this study, we present a method for real-time high-speed fire detection using deep learning. A new special convolutional neural network was developed to detect fire regions using the existing YOLOv3 algorithm. Due to the fact that our real-time fire detector cameras were built on a Banana Pi M3 board, we adapted the YOLOv3 network to the board level. Firstly, we tested the latest versions of YOLO algorithms to select the appropriate algorithm and used it in our study for fire detection. The default versions of the YOLO approach have very low accuracy after training and testing in fire detection cases. We selected the YOLOv3 network to improve and use it for the successful detection and warning of fire disasters. By modifying the algorithm, we recorded the results of a rapid and high-precision detection of fire, during both day and night, irrespective of the shape and size. Another advantage is that the algorithm is capable of detecting fires that are 1 m long and 0.3 m wide at a distance of 50 m. Experimental results showed that the proposed method successfully detected fire candidate areas and achieved a seamless classification performance compared to other conventional fire detection frameworks.

scholarly journals New functions of multicriteria ship fire alarm systems

2021 ◽  
pp. 118-123
Author(s):  
В.Н. Круглеевский ◽  
В.В. Вислогузов ◽  
A.A. Таранцев ◽  
С.Н. Турусов
Keyword(s):  
Fire Detection ◽  
Electrical Cable ◽  
Comparative Tests ◽  
Fire Alarm ◽  
Alarm Signals ◽  
Detection Systems ◽  
Alarm Systems ◽  
Characteristic Features ◽  
In Fire ◽  
Marine Vessels

В настоящей статье рассматриваются вопросы развития пожарных извещателей, контролирующих появление дыма, превышение заданного значения температуры и скорости ее роста, наличие угарного газа и использующих мультикритериальные алгоритмы для оценки обоснованности сигналов тревоги. Анализируются результаты проведенных отечественными организациями сравнительных испытаний мультикритериальных и традиционных «пороговых» пожарных извещателей и возможности их применения на судах в составе систем пожарной сигнализации. Определено, что при повторении одних и тех же модельных очагов пожаров зафиксированные значения контролируемых параметров отличались незначительно. При этом для каждого модельного очага можно было обнаружить свои характерные черты. Сделан вывод о том, что внедрение мультикритериальных алгоритмов обработки информации в судовые системы обнаружения пожаров не только сокращает время обнаружения пожара, но и позволяет расширить функциональные возможности системы. Используя мультикритериальные пожарные извещатели в системах пожарной сигнализации можно будет распознавать, что именно горит: дизельное топливо, ветошь, изоляция электрического кабеля или что-либо другое. Отмечается, что требования к судовым мультикритериальным системам сигнализации обнаружения пожара нашли свое отражение в Правилах классификации и постройки морских судов Российского морского регистра судоходства. This article discusses the development of fire detectors that control the appearance of smoke, the excess of a given temperature and the rate of its growth, the presence of carbon monoxide and use multicriteria algorithms to assess the validity of alarm signals. The results of comparative tests of multicriteria and traditional fire detectors conducted by domestic organizations and the possibility of their use on ships as part of fire alarm systems are analyzed. It was determined that when the same model fires were repeated, the recorded values of the controlled parameters differed slightly. At the same time, for each model focus, it was possible to detect its own characteristic features. It is concluded that the introduction of multicriteria algorithms for information processing in ship fire detection systems not only reduces the time of fire detection, but also allows you to expand the functionality of the system. Using multi-criteria fire detectors in fire alarm systems,it will be possible to recognize what exactly is burning: diesel fuel, rags, electrical cable insulation, or anything else. It is noted that the requirements for ship multicriteria fire detection alarm systems are reflected in the Rules for the Classification and Construction of Marine Vessels of the Russian Maritime Register of Shipping.

scholarly journals Automatic Fire and Smoke Detection Method for Surveillance Systems Based on Dilated CNNs

Atmosphere ◽  
2020 ◽  
Vol 11 (11) ◽  
pp. 1241
Author(s):  
Yakhyokhuja Valikhujaev ◽  
Akmalbek Abdusalomov ◽  
Young Im Cho
Keyword(s):  
Cost Effective ◽  
Classification Performance ◽  
Detection Methods ◽  
False Alarms ◽  
Surveillance Systems ◽  
Smoke Detection ◽  
Detection Systems ◽  
Number Of False Alarms ◽  
Unseen Data ◽  
Vision Detection

The technologies underlying fire and smoke detection systems play a crucial role in ensuring and delivering optimal performance in modern surveillance environments. In fact, fire can cause significant damage to lives and properties. Considering that the majority of cities have already installed camera-monitoring systems, this encouraged us to take advantage of the availability of these systems to develop cost-effective vision detection methods. However, this is a complex vision detection task from the perspective of deformations, unusual camera angles and viewpoints, and seasonal changes. To overcome these limitations, we propose a new method based on a deep learning approach, which uses a convolutional neural network that employs dilated convolutions. We evaluated our method by training and testing it on our custom-built dataset, which consists of images of fire and smoke that we collected from the internet and labeled manually. The performance of our method was compared with that of methods based on well-known state-of-the-art architectures. Our experimental results indicate that the classification performance and complexity of our method are superior. In addition, our method is designed to be well generalized for unseen data, which offers effective generalization and reduces the number of false alarms.

Implementation of the JPEG on DSP Processors

2010 ◽  
Vol 34-35 ◽  
pp. 1536-1539 ◽  
Author(s):  
Shi Jun Li ◽  
Xi Long Qu ◽  
Qiang Li
Keyword(s):  
Image Compression ◽  
Real Time ◽  
High Speed ◽  
International Standards ◽  
Surveillance Systems ◽  
Operation Performance ◽  
Real Time Processing ◽  
Dct Transform ◽  
Digital Image Compression ◽  
Dsp Processors

this paper introduces the design and implementation of JPEG image compression based on the high speed DSP TMS320VC5416 available from Texas Instruments. Especially, the realization and optimization of DCT transform is discussed and the image Lena is compressed with different way. Experiments show that the reconstructed images have PSNR above 34dB . JPEG algorithm is a digital image compression algorithm with high compression ratio, little distortion characteristics, and has been identified as international standards. This standard has been widely used in digital cameras, surveillance systems, mobile phones, video phones, and many other aspects. It is important to research and realize a real-time image compress system Using JPEG. DSP is used in real-time processing and portable applications with special hardware structure. DSP with high processing speed and excellent operation performance is particularly adapted to image processing. This article introduces DSP-based implementation of JPEG[1].

scholarly journals Two-Step Real-Time Night-Time Fire Detection in an Urban Environment Using Static ELASTIC-YOLOv3 and Temporal Fire-Tube

Sensors ◽  
2020 ◽  
Vol 20 (8) ◽  
pp. 2202 ◽  
Author(s):  
MinJi Park ◽  
Byoung Chul Ko
Keyword(s):  
Urban Areas ◽  
State Of The Art ◽  
Fire Detection ◽  
Classification Performance ◽  
Performance Comparison ◽  
Detection Methods ◽  
Night Time ◽  
Detection Systems ◽  
Ambient Lighting ◽  
Fast Classification

While the number of casualties and amount of property damage caused by fires in urban areas are increasing each year, studies on their automatic detection have not maintained pace with the scale of such fire damage. Camera-based fire detection systems have numerous advantages over conventional sensor-based methods, but most research in this area has been limited to daytime use. However, night-time fire detection in urban areas is more difficult to achieve than daytime detection owing to the presence of ambient lighting such as headlights, neon signs, and streetlights. Therefore, in this study, we propose an algorithm that can quickly detect a fire at night in urban areas by reflecting its night-time characteristics. It is termed ELASTIC-YOLOv3 (which is an improvement over the existing YOLOv3) to detect fire candidate areas quickly and accurately, regardless of the size of the fire during the pre-processing stage. To reflect the dynamic characteristics of a night-time flame, N frames are accumulated to create a temporal fire-tube, and a histogram of the optical flow of the flame is extracted from the fire-tube and converted into a bag-of-features (BoF) histogram. The BoF is then applied to a random forest classifier, which achieves a fast classification and high classification performance of the tabular features to verify a fire candidate. Based on a performance comparison against a few other state-of-the-art fire detection methods, the proposed method can increase the fire detection at night compared to deep neural network (DNN)-based methods and achieves a reduced processing time without any loss in accuracy.

scholarly journals High-speed algorithm for transmitting video information about emergency situations on transport objects

2021 ◽  
pp. 64-70
Author(s):  
И.Г. Малыгин ◽  
О.А. Королев
Keyword(s):  
Real Time ◽  
High Speed ◽  
High Performance ◽  
Low Complexity ◽  
Transformation Method ◽  
Video Stream ◽  
Surveillance Systems ◽  
Video Information ◽  
Emergency Situations ◽  
Multiplication Operation

Современные интеллектуальные видеосистемы наблюдения стали все больше акцентироваться на передачу в реальном времени высококачественного видео различных важных событий, в том числе чрезвычайных ситуаций. Для высокопроизводительных систем передачи видеоинформации нового поколения необходимы эффективные структурные решения, способные как к высокой скорости передачи, так и к высокой точности вычисления. Такие структуры должны обрабатывать огромные последовательности изображений, при этом каждый видеопоток должен характеризоваться высоким разрешением с минимальным шумом и искажениями, потребляя при этом как можно меньше мощности. Спектральные алгоритмы обработки видеоинформации являются наиболее распространенным способом передачи в реальном времени, в частности дискретное косинусное преобразование. При этом исходное изображение подвергается преобразованию из пространственной в частотную область с целью сжатия путём уменьшения или устранения избыточности визуальных данных. Неявное вычисление преобразования последовательности 8-точечного массива приводит к эффективному сжатию, требующему не более пятикратного выполнения операции умножения. В статье предложены архитектура с низкой структурой сложности и метод преобразования изображений на основе алгебры целых чисел. Modern intelligent video surveillance systems have become increasingly focused on real-time transmission of high-quality video of various important events, including emergencies. For high-performance video information transmission systems of the new generation, efficient structural solutions are needed that are capable of both high transmission speed and high calculation accuracy. Such structures must process huge sequences of images, and each video stream must be characterized by high resolution and with minimal noise and distortion, while consuming as little power as possible. Spectral algorithms for processing video information are the most common method of transmission in real time, in particular the discrete cosine transform. In this case, the original image is transformed from the spatial to the frequency domain in order to compress by reducing or eliminating the redundancy of visual data. Implicitly calculating the sequence transformation of an 8-point array results in efficient compression, requiring no more than five times the multiplication operation. In this paper, we propose an architecture with a low complexity structure and image transformation method based on the algebra of integers

scholarly journals Fire Detection Using Multi Criteria Image Processing Technique in Video Sequences

2015 ◽  
Vol 16 (1) ◽  
pp. 136
Author(s):  
Behrouz Memarzadeh ◽  
Mohammad Ali Mohammadi
Keyword(s):  
High Speed ◽  
Fire Detection ◽  
Detection Algorithm ◽  
Processing Technique ◽  
Video Data ◽  
Image Processing Technique ◽  
Video Sequences ◽  
False Alarms ◽  
Surveillance Systems ◽  
Gray Level

Vision-based flame detection has drawn significant attention in the past decade with camera surveillance systems becoming ubiquitous. This paper proposes a multi criterion method to detect fire or flames by processing the video data generated by a high speed camera. Since flame images are special class of images, some of the unique features of a flame may be used to identify flame. There are some differences between flame images and other general images. By using these features we are able to detect fire correctly with least false alarm. In this paper we present an algorithm which can detect fire and reduce number of false alarms by counting number of identified pixels. In the algorithm, we preprocess the images to have better results. So first we adjust the gray level of a flame image according to its statistical distribution to have better processing. After that we try to extract fire features in images. First by using color characteristics, the ratio of red to green, we can identify probable fire-like or fire like pixels. Second, to highlight the regions with high gray level contrast at their edges, we use the extended prewitt filter. We use AND operation on two above processing images to remove unrelated pixels, at last by using flicker frequency, the oscillating change in the number of identified pixels over time is transformed into the frequency domain to complete detection algorithm. Simulation proves the algorithm ability to detect fire in different situations in video sequences.

An approach for the compensation of the background variability in fire detection systems

2001 ◽  
Vol 42 (1) ◽  
pp. 23-30 ◽  
Author(s):  
M.F. Ugarte ◽  
R.I. Zequeira ◽  
F. López
Keyword(s):  
Fire Detection ◽  
Detection Systems ◽  
In Fire

scholarly journals Decision algorithms in fire detection systems

2011 ◽  
Vol 8 (2) ◽  
pp. 155-161 ◽  
Author(s):  
Jovan Ristic ◽  
Dragana Radosavljevic
Keyword(s):  
False Alarm ◽  
Fire Detection ◽  
Working Time ◽  
Detection Systems ◽  
In Fire ◽  
Decision Algorithms

Analogue (and addressable) fire detection systems enables a new quality in improving sensitivity to real fires and reducing susceptibility to nuisance alarm sources. Different decision algorithms types were developed with intention to improve sensitivity and reduce false alarm occurrence. At the beginning, it was free alarm level adjustment based on preset level. Majority of multi-criteria decision work was based on multi-sensor (multi-signature) decision algorithms - using different type of sensors on the same location or, rather, using different aspects (level and rise) of one sensor measured value. Our idea is to improve sensitivity and reduce false alarm occurrence by forming groups of sensors that work in similar conditions (same world side in the building, same or similar technology or working time). Original multi-criteria decision algorithms based on level, rise and difference of level and rise from group average are discussed in this paper.

Sign in / Sign up

Export Citation Format

Share Document