This paper describes a technique for finding regular grids in the images of calibration patterns, a crucial step in calibrating cameras. Corner features located by a corner detector are connected using Delaunay triangulation. Pairs of neighboring triangles are combined into quadrilaterals, which are then topologically filtered and ordered. We introduce a unique data structure for representing both triangular and quadrilateral meshes. This mesh structure allows us to exploit the strong topological constraints in a regular grid. Experiments show that the method is able to handle images with severe radial distortions. Implemented on a conventional desktop, grid matching can be done in real time. The method is also applicable to marker detections for augmented reality and robot navigation.