/NTL-Asteroid-Data-Hunter

NASA's Asteroid Data Hunter challenge tasked competitors to develop significantly improved algorithms to identify asteroids in images from ground-based telescopes. This winning solution increased the detection sensitivity, minimize the number of false positives, ignore imperfections in the data, and run effectively on all computers.

Primary LanguageC++Apache License 2.0Apache-2.0

NASA's Asteroid Data Hunter

Using the latest developments in machine learning algorithms, we have developed an algorithm that is capable of utilizing imagery data from modern telescopes to find more asteroids than has previously ever been possible.

This new method is approximately 15% better than the current method of identifying asteroids in the main belt (Asteroids that orbit between Mars & Jupiter). The algorithm is capable of running on a common laptop/desktop. Algorithms like these will be used on future spacecraft to identify asteroids to maximize the capability of missions in the future.

The application contains a user interface that anyone can use without too much of a learning curve for new users. It’s also easy to install and comes with a one-click installation process (no configuration necessary!). For the expert user, full documentation and source code are available for modifications and tweaking.

Learn more about the project at: http://www.topcoder.com/asteroids

NOTE: If you run into any issues, please do not hesitate to contact Rashid Sial (rsial@topcoder.com) for support, or to log a ticket in our GitHub repository.

Thanks! -The NASA Tournament Lab