一种遥感影像机场ROI检测方法  

A Detecting Approach of Airport ROI in Remote Sensing Image

作　　者：因倩[1,2] 张占睦[1] 张振[1] 

机构地区：[1]信息工程大学测绘学院 [2]61486部队

出　　处：《测绘科学技术学报》2010年第4期280-284,共5页Journal of Geomatics Science and Technology

摘　　要：提出了一种卫星遥感影像中机场兴趣区(ROI)检测的方法。该方法采用由上而下的知识驱动策略,以机场知识库为基础,根据遥感影像中机场ROI检测的原则,有针对性地选择和组织影像处理的成熟算法提取影像中的感兴趣特征;并将其与机场知识模型进行多层次的推理和判断,最终确定机场ROI影像位置并输出结果。实验表明,基于知识的机场ROI检测方法在提高检测速度,增强可靠性方面具有一定的优势,并能够有效地降低大幅面遥感影像的处理难度。A method of airport region of interest（ROI） detection in remote sensing image was proposed.It adopted the top-down knowledge driving strategy.Base on the knowledge database of airport and according to the principle of airport ROI detection in remote sensing image,some interest features in image are extracted by selecting and organizing some mature image process algorithms targetly.Then the airport region is located by multi-hierarchical judging and identifing the features with the knowledge model of airport.The experiments have demonstrated that the method of airport detection have advantages in raising processing speed and strengthen robustness.It can decrease the difficulty of processing large images effectively.

关 键 词：遥感影像 机场兴趣区 检测 知识库 特征提取 推理机制 

分 类 号：P237[天文地球—摄影测量与遥感]