Classification of Line and Character Pixels on Raster Maps Using Discrete Cosine Transformation Coefficients and Support Vector Machine

Authors:
Yao-Yi Chiang;Craig A. Knoblock
Affiliations:
University of Southern California;University of Southern California
Venue:
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02
Year:
2006

Citing 0
Cited 3

Identifying Maps on the World Wide Web

GIScience '08 Proceedings of the 5th international conference on Geographic Information Science
Automatic extraction of road intersection position, connectivity, and orientations from raster maps

Proceedings of the 16th ACM SIGSPATIAL international conference on Advances in geographic information systems
Automatic and Accurate Extraction of Road Intersections from Raster Maps

Geoinformatica

Quantified Score

Hi-index	0.00

Visualization

Abstract

Raster maps are widely available on the Internet. Valuable information such as street lines and labels, however, are all hidden in the raster format. To utilize the information, it is important to recognize the line and character pixels for further processing. This paper presents a novel algorithm using 2-D Discrete Cosine Transformation (DCT) coefficients and Support Vector Machines (SVM) to classify the pixels of lines and characters on raster maps. The experiment results show that our algorithm achieves 98% precision and 85% recall in classifying the line pixels and 83% precision and 96% recall in classifying the character pixels on a variety of raster map sources.