Kernel-Based Moving Object Detection

点击翻译按钮获取中文摘要

This award supports funding to work on a new search algorithm for moving faint objects and apply it to existing datasets in the Sloan Digital Sky Survey II (SDSS-II) and Dark Energy Survey (DES) in order to discover new objects in our solar system. The new design, called Kernel-Based Moving Object Detection (KB-MOD) will speed computations by a factor of a hundred, and use a probabilistic framework to help the user make better decisions about when a new object has been discovered, and when the limitations are reached and one should stop looking. The improved sensitivity will make it possible to discover smaller and more distant (and hence, fainter) outer solar system objects, which will help complete the census of the outer solar system, including the Kuiper Belt region, beyond Neptune. The PI and his team will train and mentor students from traditionally underrepresented groups through the University of Washington's Pre-Major in Astronomy Program (Pre-MAP), which targets students who have not yet declared a major and integrates them into the research environment so that they will be more likely to major in science. The scope of the proposal includes developing a probabilistic, image-based analysis infrastructure, and build and implement scalable methods to access the data using trajectories from a wide range of potential solar system orbits. The work will also integrate science data from the SDSS and DES into the LSST Data Management environment.