Contour line extraction from a topographical map is an integral part of Geographical Information System. Traditional technique for contour line extraction include manual digitization, this is a time consuming, tedious and a costly process. In addition the effectiveness of the result highly relies on the ability and skill of the researcher responsible for the same. To aid research initiatives these extraction process can be fully automated. In a typical topographical map in addition to the contours there are several other features represented such as point features (for e. g. Location), linear features (for e. g. river, road, contour etc. ) and area (for e. g. reserved forest) features to name a few. These identifiable features are represented in different color codes for ease of visual interpretation. In order to extract these features of interest, color based segmentation process can be implemented and the basis for the segmentation can be set as the digital value referring the color in which the object is represented. Firstly, Segmentation process highly depends on the ability of the researcher in selecting the digital number representing a feature i. e. selection of improper range leads to improper classification. Secondly, Segmentation process generally fails to classify those digital numbers that are located at the point of overlapping of features such as a river running across the contour. Both these causes would lead to generation broken features. Knowledge based efficient reconnection techniques can be devised for overcoming the same. This research work aims at developing various techniques for reconnecting broken contour lines and compare their performance based on their ability to determine the search space for locating the extension point and then its ability to reconnect.

Topographic map Contour Binary image Reconstruction Leech Water flow Wiper