Space target classification method based on multi-layer adversarial network

Abstract

The invention discloses a space target classification method based on a multilayer adversarial network, and the method comprises the following steps: 1, carrying out the sequential multi-frame projection of an imaging target through an imaging load, and obtaining a space image; 2, performing feature augmentation on the space image to obtain an augmented space image; 3, based on the top-layer adversarial network, performing foreground extraction on the augmented space image, and extracting foreground information; and step 4, based on conditional convolution adversarial training, performing progressive two-stage adversarial on the foreground information, extracting a general representation vector, and finishing space target classification. According to the invention, the classification problems that various types of targets exist in an imaging load acquisition image under a complex state and relative motion exists can be solved; reliable feature extraction can be achieved through gamingbased on the adversarial network, accurate classification of various complex targets is achieved, targeted application is developed, the on-orbit autonomous operation capacity of the active spacecraftis effectively improved, and important guarantee is provided for autonomous navigation and safe survival of the spacecraft.

Description

technical field [0001] The invention relates to the technical field of spacecraft navigation, in particular to a space target classification method based on a multi-layer confrontation network. Background technique [0002] With the development of space technology in various countries, the requirements for autonomy and functionality of space equipment are gradually increasing. On the one hand, for the autonomy of spacecraft, fully autonomous orbit determination is the most basic ability to survive and operate. In the process of orbit determination, images of celestial bodies such as stars and planets are important information sources. On the other hand, for the functionality of a spacecraft, whether it is safe survival or debris removal, active detection and active identification of targets are prerequisites for safe operation and mission execution. [0003] At the same time, due to the possible relative movement of orbit and attitude between the unknown space target and th...

AI Technical Summary

Problems solved by technology

This system aims to solve the classification problem of various types of targets and relative motion in the images acquired by imaging loads under complex conditions. Based on the adversarial network, reliable feature extraction can be achieved through games, and a method for the brightness, identity, and individuality of targets is proposed. Features, etc. have strong robustness, are widely applicable to the space environment, and can realize accurate classification methods for various complex targets, enabling spacecraft to achieve effective classification of payload observation images, and carry out targeted applications to effectively improve the active spacecraft. The autonomous operation capability in orbit provides an important guarantee for the autonomous navigation and safe survival of spacecraft

Images