项目名称: 基于光谱的天基空间点目标特征提取与识别
项目编号: No.61308101
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 无线电电子学、电信技术
项目作者: 孙成明
作者单位: 中国科学院光电研究院
项目金额: 24万元
中文摘要: 天基空间目标观测时,对于远距离目标,通常只能得到一个点的相关信息,包括点的位置和灰度等,损失了目标的材料、构型和姿态等特征。而代表物体固有属性的光谱差异信息可以作为识别目标的一种重要手段。目前,基于光谱数据同时提取目标的材料、构型、姿态等特征还缺乏具有针对性的、行之有效的方法,还有待深入研究。 本项目首先从目标光谱数据的产生特点出发,综合考虑目标的背景特性、材料特性、结构特性、轨道特性等因素,建立目标光谱特性的数学模型,测量并建立目标表面材料光谱特性数据样本;其次根据应用环境与目标的变化,针对性提出基于光谱数据曲线拟合、特征判定配合数据库搜索及基于光谱特性数学模型反演计算的目标特征提取与识别方法;最后进行典型参数条件下目标特征提取与识别仿真分析,为目标识别方法的验证提供依据。通过理论建模、实验测量、仿真分析探索出一套基于光谱的空间点目标特征提取与识别方法,为天基应用提供技术支撑。
中文关键词: 空间目标;光谱特性;双向反射分布函数;特征提取与识别;
英文摘要: The location and intensity of non-resolved space object can be obtained in space-based surveillance, however, the object material, configuration and attitude are lost. Spectrum difference reflecting the inherent properties of the object can be used as an important means of object recognition. At present, extraction of the object material, configuration and attitude based on the spectral data still lacks the directed and effective methodology. Recognition of non-resolved space object needs further research. In this research, firstly, considering the background, material, structure and orbit, mathematical model of spectral properties of space object will be established, and spectral properties of the object surface materials will be measured. Secondly, methods of object feature extraction and recognition will be proposed, including curve fitting, characteristic matching, database searching based on the spectral data and inverse calculation based on the mathematical model of spectral properties. Finally, simulation of object feature extraction and recognition will be done based on the above for method validation. A methodology for the material, configuration and attitude extraction and recognition of non-resolved space object from space-based spectral data will be put forward by theoretical modeling, experimental m
英文关键词: Space Object;Spectral Property;Bidirectional Reflectance Distribution Function;Feature Extraction and Recognition;