Robust affine registration method using line/surface normals and correntropy criterion

Abstract The problem of matching point clouds is an efficient way registration, which significant for many research fields including computer vision, machine learning, and robotics. There may be linear or non-linear transformation between clouds, but determining the affine relation more challenging among cases. Various methods have been presented to overcome this in literature one them variant iterative closest (ICP) algorithm. However, traditional ICP variants are highly sensitive effects such as noises, deformations, outliers; least-square metric substituted with correntropy criterion increase robustness ICPs effects. Correntropy-based robust available use point-to-point estimate clouds. Conversely, study, a line/surface normal that examines point-to-curve point-to-plane distances employed together cloud registration problems. First, maximum measure built conditions. Then, closed-form solution maximizes similarity sets achieved 2D extended 3D registration. Finally, application procedure developed method given its performance examined through extensive experiments on sets. results highlight our can align robustly precisely than state-of-the-art literature, while time process remains at reasonable levels.