改进的SIFT匹配算法
Improved SIFT Matching Algorithm
摘要: SIFT特征描述子的高维性和复杂性,不但占用较大的内存空间,而且影响特征匹配的速度。文章采用基于特征点邻域梯度统计的思想,局部统计区域由以特征点为中心的8个同心方环分割出来的环形组成,并计算出其相应像素的梯度(模值和方向),统计出8个方向的梯度累加值,然后进行从大到小的排序,最后再进行规一化处理。建立新的描述子将原来的128维向量降低到64维,实验证明此方法在保持匹配精度的情况下提高了匹配速度。
Abstract:
The high dimension and complexity of feature descriptor of SIFT, not only occupy the memory space, but also influence the speed of feature matching. We adopt the statistic feature point’s neighbor gradient method, the local statistic area is constructed by 8 concentric square ring feature of points-centered, compute gradient of these pixels, and statistic gradient accumulated value of 8 direction, then descending sort them, at last normalize them. The new feature descriptor descend dimension of feature from 128 to 64, the proposed method can improve matching speed and keep matching precision at the same time.