基于改进SURF的实时视频拼接方法  

Real-time Video Stitching Method Based on Improved SURF

作　　者：雷飞[1] 王文学[1] 王雪丽[1] 任维[1] 

机构地区：[1]北京工业大学电子信息与控制工程学院,北京100124

出　　处：《计算机技术与发展》2015年第3期32-35,共4页Computer Technology and Development

基　　金：北京市教育科技计划面上项目(KM201210005003)

摘　　要：视频拼接在现实生活中具有十分重要的意义。文中实现的功能是实时地将多路视频进行拼接以形成一个大视野的视频。为了使系统具有良好的实时性,先用DirectShow分离视频流、提取图像帧,然后提出了一种改进的SURF算法进行图像配准,最后采用渐入渐出法进行图像融合。改进的算法在特征点提取时对范围作了限制,同时改进了原有特征描述符,这样可以大大加快提取速度。在特征匹配时采用基于最近邻和次近邻距离比值的方法进行粗匹配,然后采用RANSAC方法进行提纯和求取变换矩阵,最后在视频拼接中,不是对每一帧图像进行配准和融合,而是只对首帧图像进行图像配准,之后的每帧图像利用第一帧计算出的变换矩阵进行图像拼接,这样可以加快视频拼接的速度。实验结果表明,该方法可以实时地生成无缝清晰的宽场景视频。Video stitching has great significance in real life. In this paper,achieve the function that stitches the multi-channel videos to form a large field of view video in real-time. To make the system has good real-time feature,firstly isolate the video streaming by DirectShow and use an improved SURF algorithm for image registration. The improved SURF algorithm restricts the scope in the step of feature point extraction and improves the feature descriptor,which can greatly speed up the extraction. M atch feature points based on the nearest neighbor and second-closest neighbor distance ratio method,and then use RANSAC approach to purify and calculate the transformation matrix. Finally,only register the first frame of the video image and use the calculated transformation matrix to stitch the other frames,instead of registering and fusing each frame of image,which can greatly accelerate the speed of the video mosaic. The experimental results demonstrate that this method can generate the seamless and clear video with wide scene.

关 键 词：视频拼接 DIRECTSHOW SURF 图像配准 图像融合 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]