Sub-pixel shift compensation method for temperature-induced drift in near-infrared optical positioning systems

Abstract

Near-infrared optical tracking systems (NIOTS), key components in surgical navigation, are widely used in various surgical procedures. However, thermal expansion within the system can cause sub-pixel shift errors, reducing positioning accuracy in 3D space. To explore these thermal effects and reduce temperature-related deviations, this study analyzes NIOTS parameters across varying temperatures and examines their relationship with thermal changes. A full-field data acquisition setup was developed, and a model was established to describe the link between temperature and sub-pixel shifts. Based on this model, a z axis correction was introduced to compensate for thermal-induced shifts. Experiments show that without compensation, the average measurement was 388.9465 mm, deviating -0.2245 mm from the true value. After compensation, the average improved to 389.1133 mm with a deviation of -0.0577 mm, achieving a 74% error reduction. Furthermore, the model maintained positioning errors within 0.1 mm in all directions during heating. These results confirm that the proposed compensation approach effectively mitigates temperature-induced drift and enhances NIOTS precision, providing a practical solution for stable and accurate surgical navigation.