一种巨幅遥感影像中机场ROI检测算法

目标检测一直是遥感领域的一个难题，巨幅影像处理进一步加大了目标检测的难度。为此，提出了一种基于特征空间分析的机场ROI快速检测方法，将影像分块读取，并行处理提取特征，采用Mean-shift方法对所提取特征分析聚类，最终定位出机场ROI。实验结果表明，该ROI检测方法具有很好的稳健性、自动性和并行性，并能够有效地降低巨幅影像处理的难度。

An Approach to Airport ROI Detection in Large Remote Sensing Images

Object detection has been a difficult problem in remote sensing community for a long time.To process very large image make object detection even more difficult.A method for detecting airport ROI(Region Of Interest)is proposed, which is based on feature space analysis that can fast locate airport.The image can be read into blocks for extracting features respectively and simultaneously.The extracted features are analyzed by mean-shift method,and eventually the airport regions can be located.The proposed ROI detecting method has strong robustness,autonomy and parallelism.It can drastically decrease the difficulty of processing very large images.