高大空间建筑火灾精确定位方法  

Accurate fire location method in high and large-span space buildings

作　　者：卢英[1,2] 王慧琴[1,2] 秦立科[1] 

机构地区：[1]西安建筑科技大学信息与控制工程学院,西安710055 [2]西安建筑科技大学建筑学院,西安710055

出　　处：《吉林大学学报（工学版）》2016年第6期2067-2073,共7页Journal of Jilin University:Engineering and Technology Edition

基　　金：高等学校博士学科点专项科研基金项目(20126120110008);陕西省自然科学基础研究计划面上项目(K05074);陕西省科技新星计划项目(2016kjxx-64);西安建筑科技大学校青年基金项目(QN1429)

摘　　要：根据高大空间建筑内火灾早期自动定位的需要,基于机器视觉技术提出一种精确的高大空间建筑火灾定位方法。由于在传统摄像机非线性显式标定方法中很难精确建立其复杂的数学模型,本文提出了一种改进粒子群优化(IPSO)算法与最小二乘支持向量机(LSSVM)相结合的摄像机隐式标定方法。在平面像素坐标与立体空间坐标之间建立映射;利用IPSO算法搜索LSSVM回归模型的最优参数,提高LSSVM的收敛速度和泛化能力,并通过三维重建实现火焰的空间定位。将此方法应用于高大空间建筑火灾定位,结果证明了该方法的有效性、快速性和准确性。For the requirements of interior fire location in high and large-span space buildings, an accurate 3D fire location method is proposed. To overcome the difficulty of establishing accurate mathematical model of camera in traditional non-linear calibration, an implicit camera calibration method based on Improved Particle Swarm Optimization （IPSO） algorithm and Least Square Support Vector Machine （LSSVM） is developed. First, the mapping relationship between the plane pixel coordinates and the spatial coordinates is established. Then, IPSO algorithm is applied to search the optimum parameters of the LSSVM regression model to improve the convergence speed and generalization ability. Finally the space positioning of the fire is realized by 3D reconstruction. The proposed method has been applied in fire location in high and large-span space buildings and the results suggest that the method is effective, fast and accurate.

关 键 词：计算机应用 高大空间建筑 摄像机标定 火灾定位 粒子群优化算法 最小二乘支持向量机 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]