基于矢量数据的图像分割算法的地理国情监测技术研究

Research on Geographic Monitoring Technology of Image Segmentation Algorithm Based on Vector Data

在地理国情监测中充分整合矢量数据与高分辨率遥感影像,提出一种基于矢量数据的图像分割算法,对两个不同时相均匀和连续的高分辨率遥感影像进行分割,采用基于最近邻算法的分类后比较法(PCC)和基于差熵算法的图像对象分析法(IOA)相互补充,形成完整的变化检测过程,提高了变化检测的精度,并提出一种基于矢量对象的精度评估方法。

Fully integrating vector data and high-resolution remote sensing images in geographical national condition monitoring,an image segmentation algorithm based on vector data is proposed to segment two high-scoring,uniform and continuous remote sensing images.Using post-classification comparison method(PCC)based on nearest neighbor algorithm and image object analysis method(IOA)based on difference entropy algorithm complement each other to form a complete change detection process,improve the accuracy of change detection,and propose a precision evaluation method based on vector object.