Extraction of Korean Information from Maps by Spatial Filtering Neural Network


The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 5, No. 1, pp. 223-235, Jan. 1998
10.3745/KIPSTE.1998.5.1.223,   PDF Download:

Abstract

An important aspect of image processing research is the design of an intelligent system capable of understanding geographical data presented as a paper map and of answering to user queries on spatial data. Specially, data in a typical map consist of roads, rivers, geographical boundaries, etc, so the creation of a conventional relational database containing all possible spatial and structural relationships among entities in a map is extremely difficult. In this paper, a novel approach for the extraction and recognition of character information from maps is proposed using spatial filtering neural network that refers to bionic optical neural network. The system is applied to 1/5,000 scaled maps published by the Geographical Survey Institute of Ministry of Construction of Korea. It will be shown that the system can extract character information for the various areas of maps successfully.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
L. W. Beom, J. J. Wook, H. H. Jung, K. W. Hyun, "Extraction of Korean Information from Maps by Spatial Filtering Neural Network," The Transactions of the Korea Information Processing Society (1994 ~ 2000), vol. 5, no. 1, pp. 223-235, 1998. DOI: 10.3745/KIPSTE.1998.5.1.223.

[ACM Style]
Lee Woo Beom, Joung Ji Wook, Hwang Ha Jung, and Kim Wook Hyun. 1998. Extraction of Korean Information from Maps by Spatial Filtering Neural Network. The Transactions of the Korea Information Processing Society (1994 ~ 2000), 5, 1, (1998), 223-235. DOI: 10.3745/KIPSTE.1998.5.1.223.