Precise isotropic scaling iterative closest point algorithm based on corner points for shape registration

Shaoyi Du, Wenting Cui, Xuetao Zhang, Liyang Wu, Lei Xiong

Abstract: The traditional iterative closest point (ICP) algorithm could register two points sets well, but it is easily affected by local dissimilar. To deal with this problem, this paper proposes an isotropic scaling ICP algorithm with corner point constraint. First, an objective function is proposed under the guidance of the corner points, as the corner points can preserve the similar of the whole shapes. Secondly, a new ICP algorithm is used to complete the isotropic scaling registration. At each step of this new algorithm, the correspondence is built based on the closest point searching, and then a closed-form solution of the transformation is computed. The experimental results demonstrate that our algorithm can prevent the influence of the local dissimilar and improve the registration precision compared with the traditional ICP algorithm.