Harris-Laplace结合SURF的遥感图像匹配拼接方法

Remote Sensing Image Matching and Mosaic Based on Harris-Laplace Combined with Scale-Invariant Algorithm

在遥感图像的处理和应用中,为了更好地解译、分析和研究图像信息,往往需要把两幅或多幅遥感图像拼接为一幅图像,文章针对遥感图像在旋转及受噪声影响时匹配困难的问题,提出了将Harris-Laplace(哈里斯-拉普拉斯)检测和SURF(快速稳健特征)算法相结合的匹配拼接方法。利用Harris-Laplace算法对遥感图像进行多尺度特征点检测,该特征点对光照变化、图像噪声和尺度改变具有不变性;然后,利用SURF算法确定特征点主方向并对特征进行描述;使用比值法进行初始匹配,接着用RANSAC(随机抽样一致性)算法剔除错误匹配点,并对匹配的图像进行拼接。试验结果表明,文中方法不但具有很好的抗旋转性能和抗噪声性能,而且较经典的SIFT(尺度不变特征变换)算法提高了匹配效率,能够为遥感图像的实时配准拼接以及几何定位精度评价提供有力的技术支持。

In the area of remote sensing image application, two or more images are usually mosaiced as one image. According to remote sensing image matching, a method of image matching and mosaic based on Harris-Laplace combined with SURF algorithm is proposed in this paper. Firstly, feature points are detected by using Harris-Laplace in multiple scales, which has the capability of invariance to illumination changes, image noise and scale changes. Then, by calculating with SURF algorithm, the main directions of the feature points are determined and the feature descriptors are generated. Ratio method is used to get initial matching, and RANSAC algorithm is used to eliminate errors and achieve accurate matching, then the image mosaicing completed. The experiment results show that the method proposed has good anti-rotation and anti-noise performance, and improve the matching efficiency obviously compared with the classical SIFT algorithm. The method can be well applied in the remote sensing image processing and geometric positioning accuracy evaluation.