地面景物红外成像特性  

Infrared imaging characterization of ground scene

作　　者：张正辉[1] 许士文[1] 

机构地区：[1]哈尔滨工业大学空间光学工程中心

出　　处：《红外与激光工程》2008年第4期583-587,共5页Infrared and Laser Engineering

基　　金：武器装备预研基金(9140A09010108HT0148)

摘　　要：对不同类型场景,在不同时段、气象条件下红外成像特性及其变化规律的研究,是红外图像匹配、红外成像精确制导技术研究的前提和依据。其目的是寻找适合匹配的场景、时段和对应的预处理方法。通过对同一地区、不同季节、时段大量试验数据的统计分析,使用图像灰度均值、标准差、边缘概率等几项指标,对不同场景红外成像特性的变化规律进行了研究,对其成像稳定性进行了评价。利用去均值归一化自相关算法进行的红外图像与其对应可见光边缘图像匹配,试验结果表明:利用该方法选择匹配场景,可以有效地提高红外景象匹配区选择的准确性和图像匹配概率。The study of infrared imaging characterization for the ground scene in different time and meteorological condition is the prerequisite and foundation of infrared image matching and precision guided technology. In order to find out the proper time,processing method and optimum scene which were suitable for image matching,a great deal of images from different season and time at same region were analyzed,and a new method to measure the variation of IR scene by some statistical parameters were presented,such as mean value,standard deviation,and probability of edge（POE） . The method was designed to quantitatively describe different scene properties,which were used to evaluate the scene stability of MWIR imaging. The MWIR and corresponding visible image matching was completed by NPROD. The results show that the method can correctly choose reference area and increase the probability of image matching.

关 键 词：红外成像 目标特性 景象匹配 统计分析 

分 类 号：TN249[电子电信—物理电子学]