文章基本信息

标题：Text Extraction from Raster Maps Using Color Space Quantization
本地全文：下载
作者：Sanaz Hadipour Abkenar ; Alireza Ahmadyfard
期刊名称：Computer Science & Information Technology
电子版ISSN：2231-5403
出版年度：2017
卷号：7
期号：2
页码：77-86
DOI：10.5121/csit.2017.70208
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Maps convey valuable information by relating names to their positions. In this paper we presenta new method for text extraction from raster maps using color space quantization. Previously,most researches in this field were focused on Latin texts and the results for Persian or Arabictexts were poor. In our proposed method we use a Mean-Shift algorithm with proper parameteradjustment and consequently, we apply color transformation to make the maps ready for KMeansalgorithm which quantizes the colors in maps to six levels. By comparing to a thresholdthe text layer candidates are then limited to three. The best layer can afterwards be chosen byuser. This method is independent of font size, direction and the color of the text and can findboth Latin and Persian/Arabic texts in maps. Experimental results show a significantimprovement in Persian text extraction.
关键词：Color space conversion; K-Means clustering; Mean-Shift algorithm; Quantization; Text extraction