摘要
全景图像拼接技术即通过将部分重叠区域的图像合成以描述某个场景信息的360度圆形图像.引用一种新型的基于SIFT(尺度不变特征变换)特征匹配的图像排序算法,实现图像的有序排列,针对图像拼接存在的误匹配点较多、耗时较长等问题,结合FAST算法进行特征点提取,接着针对相邻有序图像间的亮度差异采用自动校正操作,削弱了相邻图像间的亮度差异,并结合改进的Ransac算法剔除误匹配点对,最后用加权平衡算法实现图像的快速融合.实验结果表明该优化排序算法稳定、高效.
Panoramic image mosaic technology is a 360-degree circular image that combines the images of some overlapping areas to describe a particular scene. In this paper, a novel image sorting algorithm based on SIFT（Scale Invariant Feature Transform） feature matching is proposed to achieve the orderly arrangement of images. Aiming at the problem that the image mosaic has many mismatch points and takes a long time,the FAST algorithm is used to extract feature points. And then, an auto-correcting operation is adopted for the brightness difference between the adjacent ordered images to reduce the brightness difference, and the modified Ransac algorithm is used to eliminate the mismatching points. Finally, the Weighted Equalization Algorithm is used to realize the Fast fusion of the image. The experimental results show that the algorithm is stable and efficient.
出处
《计算机系统应用》
2017年第7期227-231,共5页
Computer Systems & Applications
关键词
全景图像拼接
图像排序
自动校正
加权平衡
融合
panoramic image stitching
image sorting
automatic calibration
weighted balance
fusion
