摘要
为获取宽视场的场景图像,提出了一种利用仿射光流场进行视频图像镶嵌的方法,首先根据光流约束方程,建立了关于图像间仿射变换参数的目标函数,然后通过最小化目标函数计算出图像间的仿射变换参数,再根据计算出来的仿射变换参数实现图像的配准,最后通过图像融合获取无缝镶嵌场景图像.实验表明,该方法可以有效地避免误拼接,可以获得较好的镶嵌结果.
To obtain a scene image of wide view, a video mosaicing method based on affine optical flow field is put forward. The object function on the affine transform parameters between video frames is built according to optical constraint equation. Then the affine transform parameters are calculated by minimizing the object function. According to the calculated affine transform parameters, registering images are realized. Image blending is put up to implement video mosaicing with these affine transform parameters. Experimental results using video sequence with real scene have demonstrated that the method can effectively avoid inaccurate mosaicing and obtain better mosaicing image.
出处
《南京师范大学学报(工程技术版)》
CAS
2007年第4期72-75,共4页
Journal of Nanjing Normal University(Engineering and Technology Edition)
基金
江苏省自然科学基金(kb2006218)资助项目
关键词
视频图像镶嵌
仿射光流
图像配准
video mosaicing, affine optical flow field, image registering