文件名称:jiyuSIFTtezhengmiaoshuzidelitipipeisuanfa
- 所属分类：
- 图形图像处理(光照,映射..)
- 资源属性：
- [PDF]
- 上传时间：
- 2012-11-26
- 文件大小：
- 105kb
- 下载次数：
- 0次
- 提 供 者：
- fen****
- 相关连接：
- 无
- 下载说明：
- 别用迅雷下载，失败请重下，重下不扣分！
介绍说明－－下载内容均来自于网络，请自行研究使用
针对不同摄像头的监控视频序列, 提出了一种基于视频帧 SIFT ( Scale Invariant Feature Transform, 即尺度不变特征变换)
特征跟踪的拼接方法。通过 SIFT算法提取帧图像的特征, 并在跟踪的估计区域搜索匹配特征, 从而确定待整合帧之间的变换参数。实验结果表明, 该方法较好实现视频快速拼接, 且对重叠区域小、 形变大、 有运动物体遮挡的视频具有较强的鲁棒性。-Surveillance cameras for different video sequences, a video fr a me based on SIFT (Scale Invariant Feature Transform, that is scale invariant feature transform) feature tracking on the method. fr a me by extracting SIFT features algorithm, and search tracking matching characteristics of the estimated area to be integrated to determine the transformation parameters between fr a mes. Experimental results show that the method is better for video fast stitching, and on the overlap is small, deformation, and have a block of video with moving objects is robust.
特征跟踪的拼接方法。通过 SIFT算法提取帧图像的特征, 并在跟踪的估计区域搜索匹配特征, 从而确定待整合帧之间的变换参数。实验结果表明, 该方法较好实现视频快速拼接, 且对重叠区域小、 形变大、 有运动物体遮挡的视频具有较强的鲁棒性。-Surveillance cameras for different video sequences, a video fr a me based on SIFT (Scale Invariant Feature Transform, that is scale invariant feature transform) feature tracking on the method. fr a me by extracting SIFT features algorithm, and search tracking matching characteristics of the estimated area to be integrated to determine the transformation parameters between fr a mes. Experimental results show that the method is better for video fast stitching, and on the overlap is small, deformation, and have a block of video with moving objects is robust.
(系统自动生成,下载前可以参看下载内容)
下载文件列表
jiyuSIFTtezhengmiaoshuzidelitipipeisuanfa.pdf