Fire Detection Algorithm Combined with Image Processing and Flame Emission Spectroscopy

Fire poses a significant risk to the safety, health, and property of people around the world. However, traditional ‘‘point sensor’’ fire detection techniques for indoor buildings based on air particles, air temperatures, and smoke have a low sensitivity, long response time, and poor stability. Therefore, video-based fire detection has become a particularly efficient and important method for detecting the early signs of a fire. Due to image blur, low illumination, flame-like interference and other factors, there is a certain error rate of fire recognition using video flame recognition methods. According to our previous study of a multi-feature flame recognition algorithm, a novel flame recognition algorithm based on free radical emission spectroscopy during combustion is investigated in this paper. First, multiple features are extracted from the video images by employing our proposed processing scheme. Then, the features are post-processed by a temporal smoothing algorithm to eliminate the error recognition rate, which is caused by the similar characteristics of objects between flame-like and real flame areas. In the temporal smoothing experiments, the proposed method achieves the true positive rates of 0.965 and 0.937 for butane flames and forest fire, respectively. Additionally, the spectral signals of OH, CH, C₂ and other free radicals in the combustion objects were acquired by the spectrometer. The vibrational temperature and rotational temperature are calculated after identification of the A²Δ?→?X²Π transition of the CH (410–440 nm). The flames-like are completely rejected by the proposed method in the validation experiment. In the subsequent butane combustion experiment, the vibrational temperature of the butane was 4896 K, and the rotational temperature was 2290 K. The experimental results show that real fires can be precisely recognized and that the combustion temperature can be determined from the CH emission spectroscopy. This novel method provides a new viewpoint for fire detection and recognition.     

Fire detection and 3D surface reconstruction based on stereoscopic pictures and probabilistic fuzzy logic

In this paper, a new fire detection method is proposed, which is based on using a stereo camera to calculate the distance between the camera and the fire region and to reconstruct the 3D surface of the fire front. For the purpose of fire detection, candidate fire regions are identified using generic color models and a simple background difference model. Gaussian membership functions (GMFs) for the shape, size, and motion variation of the fire are then generated, because fire regions in successive frames change constantly. These three GMFs are then applied to fuzzy logic for real-time fire verification. After segmentation of the fire regions from left and right images, feature points are extracted using a matching algorithm and their disparities are computed for distance estimation and 3D surface reconstruction. Our proposed algorithm was successfully applied to a fire video dataset and its detection performance was shown to be better than that of other methods. In addition, the distance estimation method yielded reasonable results when the fire was a short distance from the camera and the reconstruction of the 3D surface showed a shape that was almost the same as that of the real fire.     

An Improved Probabilistic Approach for Fire Detection in Videos

This paper proposes an improved probabilistic approach using two improved feature representations. These features are color and motion. First, an improved probabilistic model for color-based fire detection is proposed, and candidate fire regions are generated from this model. Then, an improved motion feature is used for final decision. The performance of the proposed approach showed about 0.2758 accuracy in false positive rate, and 0.2636 accuracy in false negative rate on a benchmark fire video database, which represents a decrease of 46.6% in false positive rate, and a decrease of 52.1% in false negative rate compared to the probabilistic approach.     

Multi-feature fusion based fast video flame detection

A video flame detection method based on the multi-feature fusion is presented in this paper. The temporal and spatial characteristics of flames, such as ordinary flame movement and color clues, a flame flickering detection algorithm is incorporated into the scheme to detect fires in color video sequences. An improved Gaussian mixture model method is firstly adopted to extract moving foreground objects from the still background of detection scenes; secondly, detected moving objects are then categorized into candidate and non-candidate flame regions by using a flame color filtering algorithm; finally, a flame flicker identification algorithm based on statistical frequency counting is used to distinguish true flames from fire-like objects in video images. Testing results show that the proposed algorithms are effective, robust and efficient. The processing rate of the flame detection method can achieve 24 fps with image size of 320 × 240 pixels on a PC with an AMD 2.04 GHz processor.     

A Saliency-Based Method for Early Smoke Detection in Video Sequences

Video-based smoke detection requires suspected smoke regions to be segmented from the complex background in the initial stage of detection. This segmentation is also important to the subsequent processes of detection. This paper proposes a novel method of segmenting a smoke region in smoke pixel classification based on saliency detection. A salient smoke detection model based on color and motion features is used. First, smoke regions are identified by enhancing the smoke color nonlinearly. The enhanced map and motion map are then used to measure saliency. Finally, the motion energy and saliency map are used to estimate the suspected smoke regions. The estimation result is regarded as our final smoke pixel segmentation result. The performance of the proposed algorithm is verified on a set of videos containing smoke. In the experiments, the method achieves average smoke segmentation precision of 93.0%, and the precision is as high as 99.0% for forest fires. The results are compared with those of three other methods used in the literature, revealing the proposed method to have both a better segmentation result and better precision. We also present encouraging results of smoke segmentation in video sequences obtained using the proposed saliency detection method. Furthermore, the proposed smoke segmentation method can be used for real-time fire detection in color video sequences.     

A Flame Detection Synthesis Algorithm

A flame detection synthesis algorithm is presented in this paper. The temporal and spatial of flames, such as flame movement, color clues and flame area variation are incorporated into the scheme to detect fires in video frames. Firstly, Choquet fuzzy integral was adopted to integrate color features and texture feature for extracting dynamic regions from video frames. Secondly, mean filtering was used to smooth RGB value of the video frame pixels and detected dynamic regions were filtered by a flame color filtering algorithm to extract candidate flame regions. Finally, a flame area variation identification algorithm was used to distinguish true flames from candidate flame regions. Experiments show that the proposed method is effective, robust and remains with strong anti-jamming performance against brightness variation. The processing rate of the flame detection method achieves 24 frames per second with image size of 320 × 240 pixels.     

Fire Detection in Video Using LMS Based Active Learning

In this paper, a video based algorithm for fire and flame detection is developed. In addition to ordinary motion and color clues, flame flicker is distinguished from motion of flame colored moving objects using Markov models. Irregular nature of flame boundaries is detected by performing temporal wavelet analysis using Hidden Markov Models as well. Color variations in fire is detected by computing the spatial wavelet transform of moving fire-colored regions. Boundary of flames are represented in wavelet domain and irregular nature of the boundaries of fire regions is also used as an indication of the flame flicker. Decisions from sub-algorithms are linearly combined using an adaptive active fusion method. The main detection algorithm is composed of four sub-algorithms (i) detection of fire colored moving objects, (ii) temporal, and (iii) spatial wavelet analysis for flicker detection and (iv) contour analysis of fire colored region boundaries. Each algorithm yields a continuous decision value as a real number in the range [−1, 1] at every image frame of a video sequence. Decision values from sub-algorithms are fused using an adaptive algorithm in which weights are updated using the least mean square (LMS) method in the training (learning) stage.     

受限空间火灾探测数据融合算法研究

为了实现对受限空间火灾发生情况多传感器自动探测信号处理,建立了基于CO浓度、烟气颗粒浓度、红外视频图像等多传感器探测的火灾探测数据融合分析系统.在受限空间内搭建基于多传感器火灾探测的数据融合分析实验平台.根据火灾红外视频图像的燃烧区域面积及圆形度等特征参数介绍了红外图像识别算法.在三种火灾探测方式特征参数的基础上提出了...     

Deep Belief Network For Smoke Detection

Forest fire is an serious hazard in many places around the world. For such threats, video-based smoke detection would be particularly important for early warning because smoke arises in any forest fire and can be seen from a long distance. This paper presents a novel and robust approach for smoke detection that employs Deep Belief Networks. The proposed method is divided into three phases. In the preprocessing phase, the region of high motion is extracted by background subtraction method. During the next phase, smoke pixel intensities are extracted from the Red, Green and Blue and Luminance; Chroma:Blue; Chroma:Red color spaces for foreground regions. Subsequently, second feature which is based on texture is computed for detecting smoke regions in which Local Extrema Co-occurrence Pattern, an improved version of local binary patterns are extracted from different foreground regions which compute not only texture of smoke but also intensity and color of smoke using Hue Saturation Value color space. Finally, Deep Belief Network is employed for classification. The proposed method proves its accuracy and robustness when tested on different varieties of scenarios whether wildfire-smoke video, hill base smoke video, indoor or outdoor smoke videos.     

An experimental study of a detector-operated water discharge system to enhance life safety

This paper presents the results of nine from a series of 51 instrumented full-scale fire tests designed to demonstrate the feasibility of using a detector-operated water discharge system to enhance the life safety of occupants of limited mobility in the room of fire origin.Tests were conducted in a room 4·0 m long × 3·6 m wide × 2·4 m high using both smouldering and flaming fires. The length of time available for escape from the room was obtained by comparing the time of detection with the earliest onset of a hazardous condition as determined by a life threat model.The system incorporated a multi-sensor, two-stage fire detection algorithm using signals from an optical smoke sensor and a heat sensor. Three sensitivity settings of the detection algorithm were assessed. It was found that at the highest sensitivity, the frequency of false alarms was likely to be unacceptable, whilst at the lowest sensitivity, the detection algorithm might fail to detect a fire before conditions within the room became a threat to life.The results show that by using the intermediate sensitivity for the detection algorithm, the system was able to respond to smouldering fires before the onset of hazardous conditions and before domestic sprinklers would operate. With fast flaming fire scenarios, the performance of the system was found to be comparable to that which would be expected from domestic sprinklers.     

Fire detection, location and heat release rate through inverse problem solution. Part II: Experiment

The design and fabrication of a prototype video fire detection system, which can locate a fire and determine its heat release rate, is described. The operation of the prototype system is demonstrated in a series of small-scale tests. The system utilizes a video camera to monitor an array of passive sensors distributed around the compartment to be protected. Each of the sensors is made up of a temperature-sensitive sheet that changes color at a prescribed temperature. In the event of an accidental fire, the plume of hot combustion gases rising from the fire will cause the temperature-sensitive sensors to be activated and change color. The times and locations of sensors changing color are used as data for an inverse problem solution algorithm, which determines the location and the heat release rate of the fire. A small-scale evaluation of the prototype video system is presented in which the prototype system is used to detect, locate and determine the heat release rate of a 2·4 kW burner placed in a 2·75 m wide by 2·75 m deep by 1·5 m high test enclosure. The accuracy of the prototype system in locating and determining the heat release rate of the small flame source placed in the reduced-scale enclosure is reported. In addition, the ability of the prototype system to make approximate measurements of the optical thickness of smoke in the enclosure, along camera-sensor lines-of-sight and then to use these measurements to locate and track the growth of a smoke plume is demonstrated.     

Automatic Detection of Deficient Video Log Images Using a Histogram Equity Index and an Adaptive Gaussian Mixture Model

Abstract: Video log images are often used by transportation agencies to manually or automatically extract roadway infrastructure information, including roadway geometry, signs, etc. Poor‐quality images, especially those having illumination‐related deficiencies caused by color corruption with a plain‐like grayscale histogram, sun glare, or darkness problems, are unacceptable and need to be identified. Manually reviewing the tens of millions of video log images for quality control is labor intensive and time‐consuming, so there is a need to develop automatic video log image quality control procedures. The contribution of this article is that it formulates a new problem of roadway video log image quality control and then proposes a reasonable solution to address this problem in the hope that it will motivate the development of new algorithms by other researchers. For the first time, an algorithm using a Histogram Equity Index (HEI) and an adaptive Gaussian Mixture Model is proposed to address the video log image quality issue by automatically detecting illumination‐related deficiencies. The Alberta Department of Transportation provided 15,489 video log images to test the proposed algorithm. Test results show that the developed algorithm can detect illumination‐related video log image deficiencies with a false positive rate of 4%, 3%, and 12%; a false negative rate of 15%, 17%, and 19% for plain‐like color corruption, dark, and sun glare conditions, respectively; computation time is 0.1 second/image. The proposed algorithm could potentially be used to improve video log image data quality control.     

基于O-YOLOv3 的无人机森林火灾识别研究

为解决传统森林火灾检测误报率高、响应速度慢等问题,提出了以无人机作为探测平台,地面站作为火灾识别系统,实现森林火灾的自动探测、识别和定位。开发了六旋翼无人机平台,通过所搭载的红外摄像机和机载计算机获取森林火灾现场图像并实时传回地面。利用地面站对所接收到的火灾图像进行处理,实现对森林火场的在线监测。在森林火灾识别算法方面,提出了O_YOLOv3 算法,采用Darknet 框架进行网络训练,使用K_means 方法自动生成锚点,有效提高火灾识别精度与响应速度。将O_YOLOv3 算法与其他几种算法进行对比实验验证本文算法的有效性。实验结果表明：O_YOLOv3 火灾识别算法能够快速、精准识别森林火灾;所研制的基于O_YOLOv3 的无人机森林火灾探测系统能够用于实际森林火灾探测。     

Early fire detection algorithm based on irregular patterns of flames and hierarchical Bayesian Networks

This paper proposes a new vision-based early fire detection method for real-world application. First, candidate fire regions are detected using a background and color model of fire. Probabilistic models of the fire are then generated based on the fact that fire pixel values in consecutive frames change constantly. These models are then applied to Bayesian Networks. This paper uses hierarchical Bayesian Networks that contain intermediate nodes. Four probability density functions for evidence at each node are used. The probability density functions for each node are modeled using the skewness of the color red, and three high frequencies obtained from a wavelet transform. The proposed system was successfully applied to various fire-detection tasks in real-world environments, and it effectively distinguished fire from fire-colored moving objects.     

Erratum to: Video Fire Smoke Detection Using Motion and Color Features

A novel video smoke detection method using both color and motion features is presented. The result of optical flow is assumed to be an approximation of motion field. Background estimation and color-based decision rule are used to determine candidate smoke regions. The Lucas Kanade optical flow algorithm is proposed to calculate the optical flow of candidate regions. And the motion features are calculated from the optical flow results and use to differentiate smoke from some other moving objects. Finally, a back-propagation neural network is used to classify the smoke features from non-fire smoke features. Experiments show that the algorithm is significant for improving the accuracy of video smoke detection and reducing false alarms.     

Video Fire Smoke Detection Using Motion and Color Features

A novel video smoke detection method using both color and motion features is presented. The result of optical flow is assumed to be an approximation of motion field. Background estimation and color-based decision rule are used to determine candidate smoke regions. The Lucas Kanade optical flow algorithm is proposed to calculate the optical flow of candidate regions. And the motion features are calculated from the optical flow results and use to differentiate smoke from some other moving objects. Finally, a back-propagation neural network is used to classify the smoke features from non-fire smoke features. Experiments show that the algorithm is significant for improving the accuracy of video smoke detection and reducing false alarms.     

Extended Rankine approach for bi-axially loaded steel columns under natural fire

Up till now, there has been limited research work conducted on bi-axially loaded steel columns under fire conditions. Under normal ambient temperature, the load-bearing capacity of steel columns is governed by the interaction of strength and stability considerations, which gives rise to the Rankine method. The authors extended this method to predict the fire resistance of steel columns subjected to bi-axial loading under standard fire curve. Basically, the authors developed an interaction equation based on failure surface to account for the effects of axial load and bending moments in two directions. Predictions from the proposed approach were benchmarked against a well-established finite element program SAFIR for steel columns under standard fire conditions. The same approach is then extended to include natural fire curves. To model a compartment fire with different geometries, thermal characteristics of boundary walls, different fire loads and ventilation factors, a zone fire modelling program Ozone was used. Coupling Ozone to SAFIR, the failure times of steel columns in a compartment fire were predicted. These numerical predictions were compared with those from the proposed approach and reasonable agreement was obtained.     

基于注意力机制的3D卷积双波段烟火识别方法

为提高复杂环境下烟火识别的精度，提出一种基于3D卷积和时空注意力机制的双波段烟火识别方法，该方法融合近红外和可见光双波段图像数据，使用视频流中基于时间的动态特征和基于空间的静态特征降低漏报率、误报率。实验结果表明，该算法在双波段数据集上的烟火识别精度达到99.90%，优于其他基于3D卷积的烟火识别算法，同时，模型具有较小的参数量，能够满足实时推理需求。因此，使用双波段特征的同时，结合注意力机制充分利用视频的动态信息，可以有效提高烟火识别精度。     

基于时间着色Petri网的建筑火灾疏散系统建模与仿真

为缓解建筑火灾疏散时间长且效率低问题,基于时间着色Petri网的理论,针对某建筑的结构特点进行建模并进行性能分析。通过算法计算与分析,合理规划不同类型人员的不同疏散通道选择,并对模型进行时间模拟仿真,提高建筑内人员疏散效率。提出带有建筑火灾疏散因素的管理疏散方法,增加相关的颜色集和时间戳,并在原模型中融入算法计算,模拟火灾疏散时间,并提出不同类别人员疏散引导的方案。达到疏散用时更短、效率更高、伤亡率更低的目标。实验结果表明,相比于其他模型,融入管理疏散算法的时间着色Petri网模型,人员平均疏散时间降低6.9 s,具有较高疏散效率。