Abstract:ABSTRACT Optically observing and monitoring moving objects, both natural and artificial, is important to human space security. Non-sidereal tracking can improve the system’s limiting magnitude for moving objects, which benefits the surveillance. However, images with non-sidereal tracking include complex background, as well as objects with different brightness and moving mode, posing a significant challenge for accurate multi-object detection in such images, especially in wide field-of-view telescope images. To achieve a higher detection precision in a higher speed, we proposed a novel object detection method, which combines the source feature extraction and the neural network. First, our method extracts object features from optical images such as centroid, shape, and flux. Then, it conducts a naive labelling based on those features to distinguish moving objects from stars. After balancing the labelled data, we employ it to train a neural network aimed at creating a classification model for point-like and streak-like objects. Ultimately, based on the neural network model’s classification outcomes, moving objects whose motion modes consistent with the tracked objects are detected via track association, while objects with different motion modes are detected using morphological statistics. The validation, based on the space objects images captured in target tracking mode with the 1-m telescope at Nanshan, Xinjiang Astronomical Observatory, demonstrates that our method achieves 94.72 per cent detection accuracy with merely 5.02 per cent false alarm rate, and a processing time of 0.66 s per frame. Consequently, our method can rapidly and accurately detect objects with different motion modes from wide-field images with non-sidereal tracking.

Development of a Tracklet Extraction Engine

Detecting Moving Objects in Photometric Images Using 3D Hough Transform

Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals

Efficient Search and Detection of Faint Moving Objects in Image Data

InterTrack: Interaction Transformer for 3D Multi-Object Tracking

Automatic Detecting and Tracking Space Debris Objects Using Active Contours from Astronomical Images

A Real-time Faint Space Debris Detector With Learning-based LCM

EasyTrack: Efficient and Compact One-stream 3D Point Clouds Tracker

Addressing single object tracking in satellite imagery through prompt-engineered solutions

Tracking Objects as Points

Weak Spatial Target Extraction Based on Small-Field Optical System

A fast algorithm for the detection of faint orbital debris tracks in optical images

Track to Detect and Segment: An Online Multi-Object Tracker

NEARBY Platform: Algorithm for Automated Asteroids Detection in Astronomical Images

Rapid automatic multiple moving objects detection method based on feature extraction from images with non-sidereal tracking

An Improved Real-Time Visual Tracking Method for Space Non-Cooperative Target

Featureless small object detection and tracking

Deep learning-assisted near-Earth asteroid tracking in astronomical images

A Robust Space Target Extraction Algorithm Based on Standardized Correlation Space Construction

Automatic Moving Object Detecting and Tracking from Astronomical CCD Image Sequences