面向小间距地图注记的智能识别研究

在充分分析地图注记特点的基础上,设计针对小间距地图注记的智能识别系统,提出分析投影积分高坡值个数的文本图像倾斜校正方法,并且,基于地图注记为印刷稿字体,其相同字号字体占位一致的特点,提出基于间距积分分割文本行图像的方法,最后利用现有小型OCR系统常用的基于结构化和统计信息相结合的特征提取方法构建特征字库,基于此特征数据库对提取的地图注记字符进行识别。

Research on Intelligent Recognition for Map Annotations with Short Gap

In this paper, first we do much analysis on map annotations, then designed an intelligent recognition system for map anno-tations with short gap and put forward a tilt correction method by analyzing the number of Projection integral with big value.Consider-ing that the map annotations are printing press type, so they should occupy the same space, this article put forward a text segmentation algorithm based on interval integral for characters with short gap.Finally, this paper used a feature detection method based on the combination of structural and statistical information, which is often applied in small OCR system, to construct a feature database for character library.And the system recognizes map annotations based on this feature database.