一种飞机目标及典型地物红外辐射仿真方法

机场场景内的飞机目标及其所处的地物背景具有重要军事应用价值,为了实现对这类目标的检测、识别以及动态监测,需要一套能够在不同季节、不同气象条件、不同时段、不同探测波段等条件下机场场景内飞机/地物红外辐射仿真的软件来提供训练样本。在Visual Studio 2010环境下,利用OpenGL构建了典型机场场景下飞机及地物背景模型,结合传热学和红外辐射理论的分析,将一款用于热红外分析的RadThermIR软件内核嵌入算法中,提出一种计算飞机及其地物背景的红外辐射场模型和构建其红外图像仿真方法。以图像灰度相似度作为评价指标,该方法和真实红外图像相比,仿真精度高于80%,实验结果表明,该方法可为实现全天候机场场景下飞机目标自动检测识别提供丰富的红外特性分析数据和特性知识训练样本。

An infrared radiation simulation method for aircrafts and typical features

Aircrafts and typical ground objects have an important application value in military. In order to obtain training samples to study the detection, recognition and dynamic monitoring of the targets, we develop a new piece of software which can achieve infrared image simulation of typical aircrafts and ground objects in different conditions, including different seasons, different weather conditions, and different times of the day. Based on the Visual Studio 2010 environment, we build a model of aircrafts and typical ground objects on the OpenGL. Besides, we adopt the theoretical analysis of heat transfer and infrared radiation, and embed the kernel of the RadThermIR which is a heat-analysis tool, into the algorithm. Finally, we propose a simulation method to calculate the infrared radiation intensity and build up infrared images of aircrafts and typical ground objects. The similarity of image grey scale is taken as the evaluation index, and the test results reveal that the simulation accuracy of infrared images can reach over 80%, providing rich training samples for target auto-detection, target recognition and object tracking of the aircrafts on airport.