Space debris detection is a key step in monitoring space debris. If space debris collides with spacecraft, it will cause equipment damage, mission failure, and so on. Therefore, the movement of debris must be monitored to achieve an effective prediction of its activities to avoid accidents.
Recently, researchers began to use image processing methods with deep learning models to detect dim and small objects. It is still a vital problem to use deep learning models to classify or detect space debris directly. Space debris is a dim and small object. The intensity of space debris is similar to that of the background noise. Besides, there are amounts of stars in the astronomical image sequences, their energy distribution and patterns are very similar to those of space debris.
A research team working with Dr. WEI Xin from the Xi'an Institute of Optics and Precision Mechanics (XIOPM) of the Chinese Academy of Sciences proposed a space debris detection method using feature learning of candidate regions in optical image sequences.
In this study, the optical image sequences were preprocessed to remove hot pixels and flicker noise and to remove the non-uniform background.
Then, stars were detected and removed, and the candidate regions of the space debris are extracted. These regions were classified by a trained deep learning model using a large number of simulated space debris with different signal to noise ratios (SNRs) and motion parameters, instead of using real space debris.
The advantage is that the model does not need to extract a sufficient number of real space debris with diverse parameters in the optical image sequences.
Finally, the correctly classified space debris was located precisely in the optical image sequences, and they were output together with the detected stars. The results were published in IEEE Access.
Diagram of space debris detection. (Image by XIOPM)
By utilizing the proposed method, the researchers tested the performance of space debris detection in optical images. It was observed that the proposed method has achieved outstanding performance.
The study provides a novel idea to combine the conventional space debris detection method with deep neural networks. By utilizing the advantages of both approaches, the research provided better performance than state-of-the-art methods. In the near future, the spacecraft will stand a safer place by monitoring the debris in space.
Detection results of space debris with different SNRs in acquired real images sequences. (Image by XIOPM)
52 Sanlihe Rd., Beijing,