基于最佳几何约束的遥感影像点匹配算法

发布时间：2018-06-13 02:47

本文选题：点匹配 + 匹配约束　；参考：《中国矿业大学学报》2017年06期

【摘要】：针对尺度不变特征变换(SIFT)点匹配算法中几何约束缺失问题,提出了一种基于最佳匹配几何约束的点匹配算法.该算法以SIFT匹配算法为基础,首先构建左右影像特征点集的转换模型,然后采用改进的量子粒子群算法对模型参数进行迭代寻优,每次粒子位置更新后,采用基于搜索圆的特征点匹配算法获取新位置下的特征点,并根据获取的特征点情况计算其适宜度与辅助适宜度来对粒子位置进行评价,经过多次迭代,最终获取匹配影像的最佳几何约束与该约束下相应的匹配点,实现了特征点的匹配.选取多幅遥感影像进行点匹配实验,结果表明:相比其它的点匹配算法,该算法在匹配点的数目与精度上都有显著提高,能够获得更好的点匹配结果.
[Abstract]:A point matching algorithm based on optimal matching geometric constraints is proposed to solve the problem of missing geometric constraints in the scale-invariant feature transformation (sift) point matching algorithm. The algorithm is based on sift matching algorithm. Firstly, the transformation model of left and right image feature points is constructed, and then the improved quantum particle swarm optimization algorithm is used to iteratively optimize the parameters of the model. The feature point matching algorithm based on search circle is used to obtain the feature points in the new position, and according to the obtained feature points, its suitability and auxiliary suitability are calculated to evaluate the particle position. Finally, the best geometric constraint of the matching image is obtained and the corresponding matching points are obtained, and the matching of the feature points is realized. Several remote sensing images are selected for point matching experiments. The results show that compared with other point matching algorithms, the algorithm can significantly improve the number and accuracy of matching points, and can obtain better point matching results.
【作者单位】：辽宁工程技术大学测绘与地理科学学院;
【基金】：国家自然科学基金项目(61540056;41401535)
【分类号】：TP751

【相似文献】