室外场景的多视角视频拼接
扩展式摄像机系统的多视角视频图像拼接因角度因素容易产生形状失真、深度失真和运动失真。为此，对现有视频图像拼接方法加入角度约束进行改进。分离视频中的运动区域与静止区域，分别采用现有方法和改进方法加以实验验证，通过对比分析，选取拼接效果最好的APAP＋角度约束方法和APAP＋SPHP＋角度约束方法进行拼接，将拼接后的静止区域和运动区域融合成宽视野的视频图像。实验结果表明，该方法能有效减少多视角视频图像拼接中的形状失真、深度失真和运动失真现象。...
Saved in:
| Published in | 计算机工程 Vol. 42; no. 4; pp. 259 - 265 |
|---|---|
| Main Author | |
| Format | Journal Article |
| Language | Chinese |
| Published |
湖南大学信息科学与工程学院,长沙410082
2016
湖南大学嵌入式系统与网络省重点实验室,长沙410082 |
| Subjects | |
| Online Access | Get full text |
| ISSN | 1000-3428 |
| DOI | 10.3969/j.issn.1000-3428.2016.04.045 |
Cover
| Summary: | 扩展式摄像机系统的多视角视频图像拼接因角度因素容易产生形状失真、深度失真和运动失真。为此，对现有视频图像拼接方法加入角度约束进行改进。分离视频中的运动区域与静止区域，分别采用现有方法和改进方法加以实验验证，通过对比分析，选取拼接效果最好的APAP＋角度约束方法和APAP＋SPHP＋角度约束方法进行拼接，将拼接后的静止区域和运动区域融合成宽视野的视频图像。实验结果表明，该方法能有效减少多视角视频图像拼接中的形状失真、深度失真和运动失真现象。 |
|---|---|
| Bibliography: | 31-1289/TP multi-view ; video stitchin; ; shape distortion ; depth distortion ： motion distortion ： an;,le constraint Aiming at the problems like shape distortion, depth distortion and motion distortion of multi-view video image stitching in an expansive camera system, this paper improves existing image stitching methods to stitch multi-angle videos by considering angle constraint. Videos are separated into moving regions and static regions, then these two regions are stitched respectively, using the existing methods and those with angle constraints. After comparing the results and analysis, the combination method of As Projective as Possible（APAP） and Shape Preserving Half Projective（SPHP） with angle constraints for static regions, and APAP with angle constraints for moving regions are selected according to their better performance. Videos are integrated with static regions and moving regions. Experimental results show that the proposed method can effectively reduce shape distortion, depth distortion and motio |
| ISSN: | 1000-3428 |
| DOI: | 10.3969/j.issn.1000-3428.2016.04.045 |