DocumentCode
2629981
Title
A method for recognizing character strings from maps using linguistic knowledge
Author
Nakamura, Akira ; Shiku, Osamu ; Anegawa, Masanori ; Nakamura, Chiaki ; Kuroda, Hideo
Author_Institution
Dept. of Electr. Eng. & Comput. Sci., Nagasaki Univ., Japan
fYear
1993
fDate
20-22 Oct 1993
Firstpage
561
Lastpage
564
Abstract
It is difficult to recognize character strings from maps, which involve various figures. The authors describe a method for recognizing character strings from these topographical maps. The method consists of a bottom-up process and a top-down process. First, in the bottom-up process, character candidates are extracted from a map. Second, in the top-down process, these character candidates are grouped into strings using linguistic knowledge of strings. Through the top-down process, character strings are recognized correctly and efficiently from low precise character candidates extracted through the bottom-up process. The method was applied to 13 maps (512×512 pixels) involving 98 strings. As a result, 95% of 98 strings were extracted correctly
Keywords
cartography; computational linguistics; feature extraction; optical character recognition; string matching; bottom-up process; character candidates; character strings; linguistic knowledge; string recognition; top-down process; topographical maps; Buildings; Carbon capture and storage; Character recognition; Control engineering; Data mining; Dictionaries; Educational institutions; Image recognition; Knowledge engineering; Roads;
fLanguage
English
Publisher
ieee
Conference_Titel
Document Analysis and Recognition, 1993., Proceedings of the Second International Conference on
Conference_Location
Tsukuba Science City
Print_ISBN
0-8186-4960-7
Type
conf
DOI
10.1109/ICDAR.1993.395673
Filename
395673
Link To Document