Automatic recognition of map symbols for GIS input module

Author

Choi, Kwan Sun ; Ahn, Dal ; Lee, Sun-Heum ; Kong, Yong Hae

Author_Institution

Sch. of Electr. & Electron. Eng., Soonchunhayng Univ., Choongnam, South Korea

Volume

2

fYear

1997

fDate

20-22 Aug 1997

Firstpage

543

Abstract

A GIS input method that automatically recognizes may symbols is proposed. First, a map image is separated into numerals, map symbols and contour lines. Then the numerals are merged into number strings and the map symbols are recognized. The map symbols are recognized using a back propagation neural network. The recognition rate is 94.44% for 11 kinds of symbols extracted from commercial maps

Keywords

backpropagation; cartography; document image processing; feature extraction; geographic information systems; neural nets; object recognition; visual databases; GIS input method; backpropagation neural network; commercial maps; contour lines; document image processing; feature extraction; geographic information system; map image processing; map symbol recognition; map symbols; number strings; numerals; recognition rate; Backpropagation; Computer science; Convergence; Data mining; Educational institutions; Geographic Information Systems; Neural networks; Productivity; Roads; Sun;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications, Computers and Signal Processing, 1997. 10 Years PACRIM 1987-1997 - Networking the Pacific Rim. 1997 IEEE Pacific Rim Conference on

Conference_Location

Victoria, BC

Print_ISBN

0-7803-3905-3

Type

conf

DOI

10.1109/PACRIM.1997.620321

Filename

620321