基于时空块协方差融合特征的火焰识别方法  

Flame recognition method based on temporal-spatial block covariance matrix blending feature

作　　者：蒋先刚[1] 张盼盼[1] 盛梅波[1] JIANG Xiangang;ZHANG Panpan;SHENG Meibo(School of Science, East China Jiaotong University, Nanchang 330013, China)

机构地区：[1]华东交通大学理学院,南昌330013

出　　处：《计算机工程与应用》2016年第17期208-214,共7页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61262031);江西省高校科技落地计划项目(No.KJLD12067)

摘　　要：提出基于视频图像的YCrCb和CMYK空间下的颜色和纹理等时空融合特征的火灾区域探测方法,将划分为时空域方块中的帧间颜色、空间纹理分布和运动属性组合成协方差描述子融合特征,通过分析矩阵中每两特征方差对应的正、负样本关联值的分布而确定特征的选择,首次提出了通过对协方差矩阵黎曼距离的变化分析来调整特征选择和组合方式。协方差特征的度量分别采用黎曼流形接地距离、对数欧式距离和用支持向量机训练的分类器进行对比实验。实验结果证明基于协方差矩阵融合特征的火灾探测系统表现出较高的识别精度和运行效率。A flame recognition method based on video image YCrCb and CMYK space’s color, texture and other spatialtemporal blending feature is proposed, it demarcates the video frames into spatial-temporal cubic block and integrates pixel’s transformed color space’s color component, spatial texture distribution and moving characteristics into integrated covariance matrix descriptor features. It first presents a feature selecting method by evaluating corrective value distribution corresponding each two correlation coefficient of a covariance matrix. In addition, it puts forwards a feature selection and feature assemble methods by analyzing changing trend of Riemannian Manifolds distance of covariance matrix. The flame features and classification methods are analyzed and compared by covariance metrics using Riemannian Manifolds distance,logarithmic Euclidean distance and support vector machine. It is demonstrated by experiments that the flame detection system based on covariance matrix hierarchical feature has higher accuracy and recognition efficiency.

关 键 词：协方差矩阵 测地线距离 火灾探测 特征选择 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]