基于数学形态学的扫描地图点状符号识别方法

¹武汉大学遥感信息工程学院,武汉市珞喻路129号430079;²武汉大学测绘遥感信息工程国家重点实验室,武汉市珞喻路129号,430079;³湖北省清江监理有限公司,宜昌市东山大道95号443002

基金项目: 湖北省地图院资助项目(2006)；中国地质大学(武汉)资助项目(2006)；湖北省清江监理有限公司资助项目(2006)

详细信息

Identifying Symbols in Scanned Map Based on Mathematical Morphology

¹School of Remote Sensing and Information Engineering,Wuhan University,129 Luoyu Road,Wuhan 430079,China;²State Key Laboratory for Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,129 Luoyu Road,Wuhan 430079,China;³Hubei Qingjiang Supervision Co.Ltd.,95 Dongshan Road,Yichang 443002,China

Funds: 湖北省地图院资助项目(2006)；中国地质大学(武汉)资助项目(2006)；湖北省清江监理有限公司资助项目(2006)

摘要: 提出了一种基于数学形态学的扫描地图符号双向模板匹配识别方法。试验结果表明,与常规模板匹配法相比,该方法不仅能良好地去除扫描地图上噪声的影响,而且能有效、可靠地识别扫描地图符号,平均正确识别率达90%以上。
- 双向模板匹配 /
- 数学形态学 /
- top-hat变换
Abstract: An approach for identifying symbols in scanned map based on mathematical morphology is proposed.First,preprocessing of scanned map is discussed,such as geometry processing,image background subtracting and contrast enhancing based on top-hat transform,thresholding etc..Then a bidirectional matching method based on mathematical morphological is used to identify symbols in the preprocessed scanned map.The experimental results indicate that the proposed method is more efficient and more reliable in identifying symbols in scanned map than the traditional template matching method;and the average correct accuracy of symbol identification is up to 90%.
- bidirectional template matching /
- mathematical morphology /
- top-hat transform