LINCC Frameworks — KBMOD
Abstract
KBMOD (Kernel Based Moving Object Detection) is a GPU-accelerated framework for the detection of slowly moving asteroids within sequences of images. KBMOD enables users to detect moving objects that are too dim to be detected in a single image without requiring source detection in any individual image, nor at any fixed image cadence. KBMOD achieves this by "shift-and-stacking" images for a range of asteroid velocities and orbits without requiring the input images to be transformed to correct for the asteroid motion. Utilizing the image differencing capabilities of the Vera C. Rubin Science Pipelines and Convolutional Neural Network based filtering techniques, we show how KBMOD has successfully identified 105 new trans-Neptunian objects from the DECam Ecliptic Exploration Project (DEEP) data. We describe the public release of v1.0 KBMOD, improvements and enhancements in the code base and plans for future development of KBMOD to enable it to scale to the volume of data that will be generated by the Rubin Observatory Legacy Survey of Space and Time (LSST).
- Publication:
-
American Astronomical Society Meeting Abstracts
- Pub Date:
- January 2023
- Bibcode:
- 2023AAS...24110507B