2021RAA....21...48Z

A filament is an important structure for studying star formation, especially intersections of filaments which are believed to be more dense than other regions. Identifying filament intersections is the first step in studying them. Current methods can only extract two-dimensional intersections without considering the velocity dimension. In this paper, we propose a method to identify three-dimensional (3D) intersections by combining Harris Corner Detection and Hough Line Transform, which achieve a precision of 98%. We apply this method for extracting intersection structures of the OMC-2/3 molecular cloud and to study its physical properties and obtain the associated PDF distribution. Results show denser gas is concentrated in those 3D intersections.