Target Image Mosaic Algorithm Based on Interest Points Matching

摘要：

In the process of standard interlaced camera capturing target image, the frame image will be obtained by integrating of the odd field image and even filed image. If the target quickly moves in the integrated process, the double shadow image will be produced. In order to resolve the problem, the traditional processing method is to discard the odd filed image or the even filed image, which will reduce the capture image resolution. A novel automatic mosaic algorithm is proposed to resolve the problem. Each double shadow image will be split into the odd filed image and the even field image, then two filed images will be complemented to two frame images, finally mosaic the middle image of two frame images. In order to resolve the problem that the corresponding pairs of interest points are difficult to be extracted, target image mosaicalgorithm based on interest points matching is proposed in the paper. The experimental results show that the algorithm can efficiently avoid the affect of mistake matching pairs of interest points, which expands the applicable scope of the traditional mosaic algorithm.

作者: Lingjia Gu Shuxu Guo Ruizhi Ren Shuang Zhang

作者单位: College of electronic science & engineering JiLin University JiLin Province, China

会议类型: 国际会议

会议名称: 2007年通信、电路与系统国际会议(2007 International Conference on Communications,Circuits and Systems Proceedings)

会议地点: 日本福冈

会议语种:英文

在线出版日期: 2007-07-11（万方平台首次上网日期，不代表论文的发表时间）