基于分水岭和形态学的图像特征提取方法

Image Feature Extraction Based on Watershed Algorithm and Morphology

毫米波的强穿透力和全天时、全天候的工作能力,使无源毫米波探测技术在安检方面具有广阔的应用前景。针对藏匿物品所获得的毫米波辐射安检图像较差的分辨率较差和边缘模糊,提出了一种基于形态学和分水岭算法的毫米波图像特征提取方法。该方法用复合结构形态学自适应滤波方法对毫米波辐射图像进行滤波处理,并对处理后的图像使用分水岭算法进行边缘检测和特征提取分析。试验的分析结果表明,提取的图像边缘特征与藏匿物品的几何特征相符合。

With the ability of all-day,all-weather and strong penetration,MMW radiation has vast application prospects in fields of security inspection.A combination of the complex morphology self-adaptive filtering algo-rithm and the watershed algorithm was proposed according to the MMW radiation image gotten from security in-spection for hidden obj ects.The MMW radiation image was processed by the filtering algorithm and the feature of the pre-processed image was extracted and analyzed based on the watershed algorithm.The experiment re-sults showed that the feature of edges extraction was consistent to the real geometric characteristic of hidden ob-j ects.