Verification-Based Approach for Automated Text and Feature Extraction from Raster-Scanned Maps
Selected Papers from the First International Workshop on Graphics Recognition, Methods and Applications
Semantic Analysis and Recognition of Raster-Scanned Color Cartographic Images
GREC '01 Selected Papers from the Fourth International Workshop on Graphics Recognition Algorithms and Applications
| Hi-index | 0.00 |
In this paper, we describe the Fine-to-Coarse Scale Method in which the knowledge of cartographic patterns into small-scale map aids to recognize the corresponding patterns into large-scale map of the same territory. This approach exploits the user's experience providing the knowledge domain in the form of the prescribed feature-attribute set. The cartographic patterns are presented in raster maps. A map is composed by thematic layers that contain cartographic patterns. The knowledge to recognize cartographic objects in raster fine scale maps is based on the information about the objects of the coarse scale map. These recognized objects are removed from the map, reaching a simplified representation of the map. Then, the rest of objects are recognized as new cartographic material in this simplified map. The goal of these representations is to obtain a GIS database.