A Method of Space Object Shape Recognition

Abstract

The invention discloses a space object shape recognition method. When aiming at the shape recognition of GEO space objects, the influence of space observation geometry on space object feature recognition is considered, and according to the actual physical and geometric meanings, the geometric change law of similar space observations is adopted. Next, when performing object shape recognition, the photometric data are first normalized at their respective center points, and then the normalized current observed photometric data and the spatial object that has a similar spatial observation geometric change law to the current observation The normalized photometric data of different shape models in the database is used for fast dynamic time warping distance calculation. The normalized photometric data only considers the influence of the shape of the spatial object on the photometric data of the spatial object, and then realizes GEO by calculating the "shape" distance. The target shape is identified, and the minimum value and mean value identification model are given, and the current observation data is compared with the historical photometric data of the target, which can realize the long-term monitoring of the state of the high-orbit target.

Description

technical field [0001] The invention relates to a method for recognizing the shape of a spatial object, which belongs to the field of computing. Background technique [0002] For GEO targets, due to the long distance, neither optics nor radar can achieve high-resolution imaging, so high-orbit targets mainly realize the inversion of target characteristics through photometric data. High-orbit targets, especially geostationary orbit targets, generally perform tasks such as communication and navigation, information relay, missile early warning, and on-orbit services. Targets of different mission types have different platform types, and determining the shape of the platform type is the key to determining the platform type. The key is to judge the target task type and evaluate the target threat degree. [0003] Based on photometric data, there are mainly two ways to judge the shape of space objects: one is the traditional shape inversion method, which realizes the shape inversion...

AI Technical Summary

Problems solved by technology

[0004] The shape inversion of GEO targets is realized based on photometric data. The traditional inversion method only considers the influence of phase angle changes on the shape inversion of space targets, and only considers the simple observation geometry. The analysis of the observation geometry is not comprehensive enough, and the observation The conditions are harsh; the shape recognition method of spatial objects based on photometric big data directly performs machine learning on the observed photometric data for GEO targets, without considering the geometric meaning in the observation, and the purpose of machine learning will not achieve the expected effect

Images