Suppression of stray light based on energy information mining

Area of Science:

• Aerospace Engineering
• Optical Sensing
• Computer Vision

Background:

• Star trackers are vital for precision aerospace missions, demanding high accuracy.
• Stray light poses a significant challenge for optical sensors in star tracking systems.
• Existing methods struggle with accurate star spot extraction in the presence of stray light.

Purpose of the Study:

• To develop an effective stray light suppression method for star trackers.
• To improve the accuracy of star spot extraction in challenging observational conditions.
• To enhance the reliability of attitude determination in aerospace applications.

Main Methods:

• Utilizing gray-level co-occurrence matrix and k-nearest neighbor algorithm for stray light type identification.
• Optimizing background estimation parameters for improved signal-to-noise ratio.
• Employing local differential encoding and Levenshtein distance filtering for star spot extraction and noise elimination.

Main Results:

• Successfully identified different types of stray light entering the optical system.
• Achieved accurate star spot extraction even under significant stray light conditions.
• Demonstrated the algorithm's effectiveness in real-world night sky observation experiments.

Conclusions:

• The proposed star energy information mining method effectively suppresses stray light.
• This algorithm enhances the robustness of star trackers for precision aerospace missions.
• Accurate star spot extraction in the presence of stray light is achievable with the developed technique.