Feature extraction algorithm of an irregular small celestial body in a weak light environment.

Area of Science:

• Planetary Science
• Computer Vision
• Astrodynamics

Background:

• Accurate crater matching is crucial for celestial body navigation and surface analysis.
• Dim imaging conditions and irregular object features pose challenges for traditional algorithms.
• Existing methods often suffer from insufficient feature extraction and high mismatch rates.

Purpose of the Study:

• To develop an optimized crater-matching algorithm for improved accuracy in celestial body imaging.
• To enhance the extraction and matching of features from images captured in low-light environments.
• To address the challenges of insufficient feature extraction and mismatching of irregular celestial objects.

Main Methods:

• Image preprocessing including bilateral filtering and improved histogram equalization for brightness and clarity enhancement.
• Scale-invariant feature extraction using the Oriented FAST and Rotated BRIEF (ORB) algorithm.
• Feature point mismatch reduction through Hamming distance screening.

Main Results:

• Significant improvement in image quality for dim and low-light celestial body images.
• Increased number of extracted feature points compared to baseline methods.
• Reduced mismatch rate of effective feature point pairs, leading to a higher overall matching rate.

Conclusions:

• The developed crater-matching algorithm effectively enhances image quality and feature extraction in challenging lighting conditions.
• The optimization significantly improves the accuracy and reliability of celestial object recognition and matching.
• This algorithm offers a robust solution for navigation and surface analysis in planetary exploration.