A Novel Technique of Skeletonization for Feature Extraction in Cadastral Maps

A cadastral map is a map showing the boundaries and ownership of land parcels. The British evolved the present cadastral system for the purposes of governance and revenue collection. Cadastral maps can be treated as base maps for all land related information. Typically, these maps store information pertaining to ownership details, land use, areal extents, type of development that has been made by the parcel owner, topological relationship with neighboring plots etc. This paper presents a novel method for the extraction of various features of cadastral maps. The extraction process works in two phases. In the first phase, boundary pixels of the cadastral map are found and skeletonization is carried out on them. In the second phase, the manipulated data obtained from the previous phase is analyzed to extract the features of the given cadastral map. The cadastral maps are generally not very accurate. But most of the available skeletonization algorithms are very sensitive to inaccuracies. In the proposed method, the skeletonization algorithm has been modified to take care of the inherent inaccuracies present in the cadastral maps.