Improved marker coordinate recognition algorithm incorporating temporal continuity of deformation

Abstract

Marker coordinate recognition (MCR) is widely used in fields such as medical character positioning, artificial intelligence control, and motion and deformation monitoring. Registration algorithms are often applied to improve the accuracy of MCR. However, external conditions such as light and temperature induce a temporal variation in the grey level during image acquisition, thereby reducing measurement accuracy. To overcome this problem, a mechanical constraint is introduced into marker point image processing, and an algorithm that incorporates the temporal continuity of deformation is proposed. A detailed derivation of this algorithm is generated, and the algorithm's performance is systematically validated through numerical and experimental tests. The results show that the proposed method yields accurate measurements accuracy but with a slight increase in computational cost. The proposed method can be used to improve deformation measurement and position accuracy.