WND-CHARM first reduces each image to a total of 2873 numerical low-level descriptors (when the ā-lā switch in the command line is turned on, which indicates that the larger set of image features should be computed). 2008), which was originally designed for automatic analysis of cell and tissue images, but also demonstrated efficacy as a general-purpose image analysis tool ( Shamir 2008 Shamir et al. The image analysis algorithm used for the automatic galaxy image classification is WND-CHARM ( Shamir et al. In Section 2 we briefly describe the algorithm, and in Section 3 the experimental results are discussed. Full compilable source code can be freely downloaded. The algorithm was originally developed for automatic analysis of cell morphology, but its general-purpose design allows it to be effective for applications outside the scope of cell biology. Here we describe a software tool that can be used for automatic classification of galaxy images. However, the bottleneck introduced by the manual analysis limits the ability of this method to provide quick analysis of massive galaxy datasets. While each volunteer can classify just a limited number of galaxies, the efficacy of the data analysis is enabled by the availability of a very large number of human observers. The galaxy images are acquired by the Sloan Digital Sky Survey (SDSS), and displayed by Galaxy Zoo as JPEG images scaled by 0.024 R p, where R p is the Petrosian radius ( Petrosian 1976) for the galaxy. 2008), allows hobbyist volunteers to log-in and manually classify galaxies via the project web site. One approach to classification of large sets of galaxy images, which was successfully adopted by the Galaxy Zoo project ( Lintott et al. This includes the need for automatic morphological classification of celestial objects that appear inside an astronomical frame. The availability of these large datasets has introduced the need for tools that can automatically analyze astronomical images. In the past several years autonomous sky surveys have been becoming increasingly important, and large datasets of astronomical images have been generated and become available by these ventures.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |