基于SIFT算法的室内全景图拼接

室内全景图像拼接采用SIFT特征点进行图像匹配与融合。由于相机镜头视野范围有限,需要多张具有重合区域不同角度图像进行拼接,以获得完整的全景图像。首先对多张原图像进行图像增强和噪声滤波的预处理,以减少特征点提取时的干扰因素;再将多张图像压入堆栈,采用SIFT算法提取每张图像的特征点;使用FLANN快速最近邻搜索包进行最近邻特征点匹配,最后进行图像融合。试验结果表明该方法能够很好地实现室内全景图像的拼接。

Indoor Panorama Splice Based on SIFT Algorithm

SIFT algorithm is used to extract feature points to realize the mosaic of indoor panoramic images. Due to the limited field of view of the camera lens,multiple images with overlapping areas and different angles need to be spliced to obtain a complete panoramic image. In order to reduce the interference factors,such as image enhancement,noise filtering and so on,the original images should be preprocessed first. And then push the multiple images into the stack. Next extract the feature points of each image by SIFT algorithm. The nearest neighbor method of FLANN is used to match feature points,and finally image fusion is carried out. The experimental results show that this method can achieve the indoor panoramic image mosaic well.