Spatiotemporal bag-of-features for early wildfire smoke detection

Wildfire smoke detection is particularly important for early warning systems, because smoke usually rises before flames arise. Therefore, this paper presents an automatic wildfire smoke detection method using computer vision and pattern recognition techniques. First, candidate blocks are identified using key-frame differences and nonparametric smoke color models to detect smoke-colored moving objects. Subsequently, three-dimensional spatiotemporal volumes are built by combining the candidate blocks in the current key-frame with the corresponding blocks in previous frames. A histogram of oriented gradient (HOG) is extracted, and a histogram of oriented optical flow (HOOF) is extracted as a temporal feature based on the fact that the direction of smoke diffusion is upward owing to thermal convection. From spatiotemporal features of training data, a visual codebook and a bag-of-features (BoF) histogram are generated using our proposed weighting scheme. For smoke verification, a random forest classifier is built during the training phase using the BoF histogram. The random forest with the BoF histogram can increase the detection accuracy performance when compared with related methods and allow smoke detection to be carried out in near real time.     

送风对中庭烟气温度的影响

通过理论计算,得出无喷淋中庭火源上的火焰气流各分区的烟气温度和速度随中庭高度变化曲线。在模型中庭内开展了一系列的送风发烟实验,并将自然状态下的实验结果与理论结果进行了比较,结果表明在紊流区两者相差不大。     

基于CruiseControl的持续集成实现方案

软件的开发过程决定着软件质量。在任何软件开发过程中都有一个重要的组成部分：得到可靠的软件创建版本。在分析传统集成模式的弊端基础上,提出采用持续集成的观点。阐明了持续集成的基础理论以及与持续集成相关的常见的误区和概念。分析了持续集成工具CruiseControl工作原理,最后提出了一种基于CruiseControl的持续集成实现方案。     

纳米粒子几何结构对膨胀型防火涂料阻燃和抑烟性能的影响

为了研究纳米粒子几何结构对膨胀型防火涂料阻燃和抑烟性能的影响,以聚磷酸铵(APP)-季戊四醇(PER)-三聚氰胺(MEL)-硼酸三聚氰胺(MB)为阻燃体系,硅丙乳液为成膜物质,分别添加球形纳米二氧化硅(Si O2)、经硅烷偶联剂KH560改性的管状碳纳米管(KH560–CNT)和片状水滑石(LDH)制备了3种水性膨胀型防火涂料。采用小室法、隧道法及模拟大板法研究了涂层的防火阻燃性能,再结合热重分析、扫描电镜分析和能谱分析,考察了涂层的热解性能和炭层结构。结果表明,球形Si O2和片状LDH能有效增强防火涂料的阻燃和抑烟性能,表现出较好的协效作用;管状KH560–CNT则会恶化防火涂料的阻燃和抑烟性能,表现出对抗作用。当纳米填料的质量分数为0.5%时,片状LDH表现出最优的协效作用,火焰传播比值为12.5,烟密度等级为14.1%。这是因为它能促进涂层在燃烧过程中形成更多的含磷和含硼交联结构,有效增强了炭层的致密性、连续性和隔热性能。     

基于自适应神经模糊推理系统的视频烟雾检测

提出一种基于自适应神经模糊推理系统的视频烟雾检测算法。从视频图像中提取烟雾特征,采用减法聚类确定模糊规则数,建立初始模糊系统。通过神经网络的自学习机制调整前提参数和结论参数,确定模糊推理规则。实验结果表明,与传统BP神经网络算法及支持向量机算法相比,该算法具有较优的ROC曲线特性。     

Modelling of light extinction by soot particles

A simplified model for the prediction of light extinction through combustion generated smoke is presented. The model builds upon existing theory and available experimental data to account for the principal factors that influence light extinction in participating media, including wavelength, primary particle size distribution and morphological structure of particle aggregation as well as multiple scattering among particles within a particle aggregation. Good agreement is demonstrated between the model predictions and experimental data in the visible and IR ranges. The model illustrates that as the mean particle size increases, the integral optical property of a soot cloud approaches that of monodisperse particles. It is postulated in the current study that the number of particles participating in multiple scattering is around 20 regardless of the real size of particle aggregation.     

Effect of smoke application to field‐grown Merlot grapevines at key phenological growth stages on wine sensory and chemical properties

Background and Aims: Smoke exposure of grapevines and development of smoke taint in wine are issues of increasing incidence and severity. There is limited understanding of the effect of phenological stage at the time of smoke exposure on taint development. The aim of this study was to demonstrate the variation in smoke uptake and taint development between and within seasons. Methods and Results: Smoke was applied to field‐grown Merlot grapevines at 12 stages of vine development over three growing seasons. Key periods of vine sensitivity to smoke taint in wine were (i) from shoots at 10 cm to full bloom (low levels of smoke taint); (ii) from berries at pea size to the onset of veraison (variable levels of smoke taint); and (iii) between 7 days post‐veraison and harvest (high levels of smoke taint). Conclusions: The severity of taint in wine varied depending on the phenological timing of grapevine smoke exposure. Taint was elevated when exposure occurred between 7 days post‐veraison and harvest. The carry‐over of smoke constituents the following season was not detectable in wine but yields were reduced. Significance of the Study: This is the first study to demonstrate the timing of smoke exposure to critically affect wine chemical and sensory characters. These effects were consistent and reproducible over three seasons.     

工业烟气黑度监测系统工程技术研究

利用远程监控技术,采集企业违规排放烟气图像,并通过计算机图像识别技术实现对排放黑烟进行自动监测,自动分析其黑度等级,为环保执法提供依据。文章是根据环境保护对烟尘排放的具体技术要求,设计了一套城市烟气黑度自动监测系统解决方案并付诸工程实施。重点叙述了烟气黑度识别技术、无线宽带网建设、图像传输技术等项目建设的关键与技术难点。系统建成后,有效解决了工业黑烟排放污染源的监控。该系统可靠,操作简便,监控直观,达到了系统预期的工程目标,此工程技术值得在智慧城市建设中推广应用。     

基于高斯扩散模型的催泪弹气溶胶烟雾非致命效能仿真

催泪弹是一种非致命防暴弹。为研究其非致命效能大小,以武警RS97-2型催泪弹为研究对象,应用高斯扩散模型研究了催泪弹气溶胶烟雾的扩散过程。分析了烟雾扩散时西埃斯(CS)催泪剂浓度和烟雾半径的变化规律。用Matlab软件编制的程序进行了仿真计算,得出以催泪弹气溶胶烟雾扩散的有效作用区域作为其非致命效能大小的衡量指标。结果表明,在风速为2m·s~(-1)时,一枚催泪剂装药为20g催泪弹的气溶胶烟雾有效作用区域可达453.7m~2。     

一种红外烟幕三维建模仿真方法

根据红外烟幕的遮蔽原理, 结合红外烟幕消光模型和扩散模型建立红外烟幕三维空间的透过率模型。针对传统烟幕遮蔽效果仿真方法的局限性, 提出一种三维空间透过率的仿真方法,并对其进行数值计算与仿真验证。研究结果表明, 探测器观察到的目标图像是多层透过率叠加的结果, 遮蔽效果优于传统方法, 对工程应用更具意义。