New iterative closest point algorithm for isotropic scaling registration of point sets with noise

This paper proposes a new probability iterative closest point (ICP) approach with bounded scale based on expectation maximization (EM) estimation for isotropic scaling registration of point sets with noise. The bounded-scale ICP algorithm can handle the case with different scales, but it could not effectively yield the alignment of point sets with noise. Aiming at improving registration precision, a Gaussian probability model is integrated into the bounded-scale registration problem, which is solved by the proposed method. This new method can be solved by the E-step and M-step. In the E-step, the one-to-one correspondence is built up between two point sets. In the M-step, the scale transformation including the rotation matrix, translation vector and scale factor is computed by singular value decomposition (SVD) method and the properties of parabola. Then, the Gaussian model is updated via the distance and variance between transformed point sets. Experimental results demonstrate the proposed method improves the performance significantly with high precision and fast speed.