Angle Estimation based on Image Pyramid and Gray Projection

摘要：

In this paper we present a new method for angle estimation between two frames of the same scene based on image pyramid transform and gray projection algorithm. By selecting proper level of pyramid sequence and correlating their row or column gray projection arrays, angle vector between frames is computed. Since low resolution image is invariable to translation and less affected by moving object, our method is more robust and accurate in practice. During the experiment, we compared our improved method to the proposed method of angle estimation, and the experimental results have shown that the method using pyramid in practical applications has higher accuracy and is more robust to the images with translation and small moving object.

关键词： Gaussian Pyramid Gray Projection Angle estimation image motion compensation

作者: Mai Ying Zhao Hong-ying Yan Lei

作者单位: Institute of Remote Sensing and Geography Information System, Peking University, Beijing 100871, P.R.China

会议类型: 国际会议

会议名称: The 31st International Congress on Imaging Science(第31届国际影像科学大会 ICIS2010)

会议地点: 北京

会议语种:英文

页码: 471-474

在线出版日期: 2010-05-12（万方平台首次上网日期，不代表论文的发表时间）