|
|
|
| Clustering Analysis of Space Object Characteristics Based on Convolutional Neural Network |
| WANG Wen-Zhu1 LI Zhi1 LAI Jia-Zhe1 FANG Yu-Qiang1 |
| 1. Space Engineering University, Beijing 101416, China |
|
|
|
|
Abstract The clustering analysis of space targets is an e ective method for supporting command and control, such as target analysis, attribute classification and identity recognition. However, the feature of the characteristics data is hard to identify, which results in the ineciency of data clustering. Firstly, the convolutional neural network is used to extract features for obtaining high-quality data. Secondly, the K-means algorithm is used for clustering the feature data. Experimental results show that the method improves the accuracy and performance of clustering analysis significantly, which can enhance the ability of analyzing space targets, and support the decisions making on the space situation command.
|
|
|
| Fund:Supported by China Postdoctoral Science Foundation (2017M623345) and National Outstanding Youth Foundation(2017-JCJQ-ZQ-005) |
|
|
|
|
|
|
2020, Vol. 6 
