基于光谱的天基空间点目标特征提取与识别

发布时间：2018-03-07 11:36

本文选题：空间目标　切入点：光谱特性　出处：《物理学报》2015年03期 　论文类型：期刊论文

【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,包括目标的位置和灰度等,损失了目标的材料、大小和状态等特征.代表物体固有属性差异的光谱特性可作为目标特征提取与识别的一种重要手段.从目标光谱特性的产生特点出发,综合考虑目标的材料特性、结构特性、背景特性、轨道特性等因素,建立了目标光谱特性的数学模型,提出了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,以环境一号卫星缩比模型为例,进行了典型参数条件下的目标特征提取与识别实验验证,实验结果验证了建模方法的正确性.
[Abstract]:When space-based space objects are observed, only one point can usually be obtained for long-range objects, including the position and gray level of the target, and the material of the target is lost. The spectral characteristics, which represent the difference of the object's intrinsic properties, can be used as an important means to extract and recognize the target's characteristics. Based on the characteristics of the target's spectral characteristics, the material and structure characteristics of the target are considered synthetically. Based on the background characteristics and orbit characteristics, a mathematical model of the spectral characteristics of the target is established, and a method of feature extraction and recognition based on the inverse calculation of the mathematical model of the spectral characteristics is proposed. The experimental results of target feature extraction and recognition under typical parameters have been carried out, and the correctness of the modeling method has been verified by the experimental results.
【作者单位】：中国科学院光电研究院;北京航空航天大学精密光机电一体化技术教育部重点实验室;
【基金】：国家自然科学基金青年科学基金(批准号:61308101) 教育部长江学者和创新团队发展计划(批准号:IRT0705)资助的课题~~
【分类号】：P171

【参考文献】