As part of the Asteroid Grand Challenge Series, NASA ran a two-phased project in 2014 to 1) develop an algorithm that cleans the output of the Catalina Sky Survey (CSS) algorithm, detecting the asteroids on space images and screen out false positive detections 2) develop an algorithm that outperforms the CSS algorithm in detecting the asteroids on space images that could be operated by a non-experienced user on a laptop.
The first phase generated 69 competitors making 301 submissions for a $10,000 prize purse (see the problem statement). The solution decreased the frequency of false positive detections by an order of magnitude. The second phase consisted of 47 competitors making 256 submissions for a $20,000 prize purse (see the problem statement). The winning algorithm completes an analysis of a set of four images in 30 seconds on a laptop and detects at least 15% more verified objects than the current state of the art. For more information, please visit here or here.
