遥感影像图像拼接算法研究
王颖;王忠庆
【期刊名称】《电子世界》 【年(卷),期】2014(000)013
【摘要】For the image features of high resolution remote sensing image,using SURF registration algorithm for mosaicing image quickly. In this scheme,SURF(Speeded Up Robust Features)is applied to extract features from the images;and using the geometric features of image to screen the feature points,in order to improve the efficiency of image mosaic and reduce the error in calculating;and then,for eliminating the matching
error,bring
in
the
RANSAC(Random
Sampling
Consensus)algorithm for image registration;for the non-uniform ilumination,lead to the seaming image interpolations are not uniform. In order to solve this problem,we using the Cubic Spline Interpolation,so that the whole image is more consistent with human eyes evaluation standard after mosaic.%针对高分辨率遥感影像的图像特点，采用SURF特征算法实现对图像的快速拼接。该方法利用SURF算子提取特定区域的图像特征；并应用图像的几何特征做进一步的特征点筛选，以提高图像拼接效率，减少计算误差；利用RANSAC配准算法对图像进行配准，以消除误匹配；对于光照不均匀情况，致使拼缝图像插值不均匀，可采用三次样条插值法处理，使拼接后图像整体更符合人眼评价标准。 【总页数】1页(92-92)
遥感影像图像拼接算法研究
