Abstract: This paper proposes an improved RANSAC algorithm based on structural similarity to improve the speed and accuracy of traditional RANSAC（Random Sample Consensus） algorithm. Firstly, BRISK（Binary Robust Invariant Scalable Keypoints） algorithm is used to detect and describe feature points. The initial match set is obtained by hamming distance feature matching. Then, false match is eliminated by structural similarity constraint. Finally, the new match set is taken as the input of RANSAC to calculate the transformation matrix. The algorithm can obtain the transformation model quickly because it has purified matched points after the initial matching. Experiments show that the number of iterations and run time are obviously less than the traditional algorithm. Therefore, the proposed algorithm outperforms the traditional RANSAC algorithm in terms of both speed and accuracy.

Key words: Random Sample Consensus（RANSAC） algorithm, feature matching, structural similarity, purify matched points

摘要： 为了减少传统RANSAC（Random Sample Consensus，随机抽样一致性）算法的迭代次数和运行时间，提高算法的速度和精度，提出了一种基于结构相似的RANSAC改进算法。采用BRISK（Binary Robust Invariant Scalable Keypoints）算法提取和描述二进制特征点，用Hamming距离进行特征匹配，获得初始匹配点集，利用结构相似约束剔除误匹配点，得到新的匹配点集，用新的点集作为RANSAC的输入，求出变换矩阵。该算法在初始匹配后进行了匹配点提纯，能快速求得变换模型。实验证明该算法迭代次数和运行时间比传统RANSAC算法明显减少，因此改进的算法在速度和精度上优于传统的RANSAC算法。

关键词: 随机抽样一致性（RANSAC）算法, 特征匹配, 结构相似, 匹配点提纯