基于先验统计知识和文字线的地名提取方法

对扫描地图地名注记的自动提取进行了研究,针对复杂背景扫描地图设计了一种新的地名注记提取方法。首先,运用先验统计知识对地名注记所在要素层进行分割;然后,将文字线组合方法与数学形态学膨胀算法结合实现了地名注记的高效提取。通过大量实验表明,该方法有效地克服了复杂背景下地名注记提取的困难,简单、快速且正确率较高。

Toponym Extraction Method Based on a Priori Statistical Knowledge and Character Line

This paper makes a study on the automatic extraction of toponym annotation in the foreign scanned map.A new method for the extraction of toponym annotation is designed in the light of the scanned map with complex background.First of all,a priori statistical knowledge is used to segment the toponym layer.Then the character line organizational method and mathematical morphological expansion algorithm is combined to achieve the effective extraction of toponym.A large number of experiments show that this method can effectively overcome the difficulty of the extraction of toponym in complex background,which is simple,rapid and has high accuracy.