基于改进RANSAC算法的图像拼接技术

Image Mosaic Method Based on Improved RANSAC

针对传统RANSAC算法在图像拼接中效率低的问题,提出了一种解决该问题的新算法,即M_RANSAC算法.该方法首先通过HARRIS算法提取2幅图像中的特征点,且在特征点匹配排序的基础之上,根据数据错误率得出抽样次数,并采用双阈值法进行数据检验来提高算法效率.结果表明,M_RANSAC算法能有效地减少抽样时间和数据检验时间,同时降低了误矩阵的估算概率,提高了图像拼接的效率.

Because of the disadvantages of the low efficiency of traditional RANSAC algorithm in image mosaic, a new algorithm, M_RANSAC algorithm, was proposed. Firstly, the feature points were extracted by HARRIS algorithm, then based on the sorting of the matching pair of feature points, the times of random selection were ob- tained by the error rate of data and the data testing were implemented by the method of double threshold. The re- suits indicated that this new algorithm, M_RANSAC algorithm, can efficiently decrease the time of the random selection and the date testing, reduce the likelihood of false matrix assessing at the same time, which can be ap- plied in image mosaic successfully.