Blind Detection of Ultra-faint Streaks with a Maximum Likelihood Method
Abstract
We have developed a maximum likelihood source detection method capable of detecting ultra-faint streaks with surface brightnesses approximately an order of magnitude fainter than the pixel level noise. Our maximum likelihood detection method is a model based approach that requires no a priori knowledge about the streak location, orientation, length, or surface brightness. This method enables discovery of typically undiscovered objects, and enables the utilization of low-cost sensors (i.e., higher-noise data). The method also easily facilitates multi-epoch co-addition. We will present the results from the application of this method to simulations, as well as real low earth orbit observations.
Keywords
Cite
@article{arxiv.1609.07158,
title = {Blind Detection of Ultra-faint Streaks with a Maximum Likelihood Method},
author = {William A. Dawson and Michael D. Schneider and Chandrika Kamath},
journal= {arXiv preprint arXiv:1609.07158},
year = {2016}
}
Comments
11 pages, 7 figures, contribution to the Advanced Maui Optical and Space Surveillance Technologies Conference 2016