Creating Cylindrical Panoramic Mosaic from a Pipeline Video through MRF based Optimization
In geological engineering, stratum structure detection is a fundamental problem affecting project planning and implementation. One of the most commonly employed detection technologies is to take videos of borehole by a forward moving camera. A coupled problem with this approach is how to construct a panoramic image from such video sequences, which are typically shot in low quality. In this paper, we propose a novel method to create a seamless panoramic image of the borehole from the video sequence without camera calibration and tracking. To stitch together pixels of neighboring frame images, our camera model is associated with a focal length changing feature, along with a small rotation possibility in the two-dimensional image space. Essentially, our camera model assumes objects lie on a cylindrical wall and the camera moves forward along the central axis of the cylindrical wall. Our method robustly resolves these two degrees-of-freedoms through KLT feature tracking. Since the quality of the video is affected by possible overflow of illumination, blurring of camera lens, and low resolution due to the distance from the objects, we build a cost function based on Markov Random Field, which is optimized using a belief propagation algorithm. Therefore, we can automatically construct the panorama with good resolution, smoothness, and continuousness in both texture and illumination space. Experiments show that our method could efficiently produce high-quality panorama for very long video sequences.
国际会议
2011国际计算机辅助设计与图形学学术会议(CAD/Graphics 2011)
济南
英文
1-8
2011-09-15（万方平台首次上网日期，不代表论文的发表时间）