于鲲, 郭彪, 丛明煜. 空间目标临边探测背景红外成像建模与图像仿真[J]. 红外与激光工程, 2019, 48(9): 904005-0904005(10). DOI: 10.3788/IRLA201948.0904005
引用本文: 于鲲, 郭彪, 丛明煜. 空间目标临边探测背景红外成像建模与图像仿真[J]. 红外与激光工程, 2019, 48(9): 904005-0904005(10). DOI: 10.3788/IRLA201948.0904005
Yu Kun, Guo Biao, Cong Mingyu. Infrared imaging modeling and image simulation of limb background for space target detection[J]. Infrared and Laser Engineering, 2019, 48(9): 904005-0904005(10). DOI: 10.3788/IRLA201948.0904005
Citation: Yu Kun, Guo Biao, Cong Mingyu. Infrared imaging modeling and image simulation of limb background for space target detection[J]. Infrared and Laser Engineering, 2019, 48(9): 904005-0904005(10). DOI: 10.3788/IRLA201948.0904005

空间目标临边探测背景红外成像建模与图像仿真

Infrared imaging modeling and image simulation of limb background for space target detection

  • 摘要: 临边红外观测是对辐射强度较弱空间目标的一种有效探测方式,探测临边背景的红外辐射分布规律直接影响目标检测跟踪的效果。通过实测图像对临边背景辐射统计特征进行建模:临边背景辐射亮度在地平线垂直和水平方向具有不同的分布规律,垂直分布与视线切线高度相关,水平方向可近似为正态分布,变异系数取值集中于0.01,0.05。基于建模成果提出了一种临边背景红外图像快速仿真方法,实验结果表明仿真图像与实测图像具有较高的一致性,验证了成像模型和仿真方法的有效性。同时,快速仿真方法耗时仅为常规逐像元仿真方法的1/200,能够满足实时仿真的需求。研究结果可应用于红外相机设计和目标检测算法测试等领域。

     

    Abstract: The limb infrared sounding is an effective means of detecting the space target with weak radiation. The infrared radiation characteristics of the limb background have an important influence on the target detection and tracking. The radiation statistical characteristics of the limb background were modeled by the measuring images:the background radiance had different distribution patterns in the vertical and horizontal directions of the horizon projection in the image. The vertical distribution was related to the tangent height. The horizontal direction could be approximated as a normal distribution, and the coefficient of variation ranged from 0.01 to 0.05. Based on the modeling results, a fast simulation method of limb background infrared image was proposed. The experimental results show that the simulation image and the measuring image have high consistency, which verifies the effectiveness of the imaging model and simulation method. The time of the fast simulation method is 1/200 of the conventional pixel-by-pixel simulation method, which can meet the requirements of real-time simulation. The research results can be applied to the fields of infrared camera design and target detection algorithm testing.

     

/

返回文章
返回