Algorithm increases effectiveness of crowdsourced astrophotography

MIT Technology Review : Digital photography and editing software have made astrophotography easier for the average person. And combining crowdsourced images of astronomical objects to create even clearer pictures could supplement professional imagery. Unfortunately, most amateur photographers do not document their camera settings and apply various filters and other changes to their pictures. A new algorithm produced by Dustin Lang of Carnegie Mellon University and his colleagues may overcome that problem. The researchers took more than 5000 pictures of NGC 5907 posted on Flickr, Bing, and Google, aligned the images using their site astrometry.net (which provides celestial coordinates for images), and then ran the images through their algorithm. The resulting images reveal details similar to those found in an 11-hour exposure from a 50-cm telescope, with many details being invisible in the individual source photos. The algorithm scales linearly based on the number of images being used, which means it could be very useful for large collections. And it could also be used for examining historical images, one of the original goals of the project.