Kalman filter based spatial-compensated 3D microscopy

Extended depth of field (EDOF) achieves high-resolution 3D microscopic measurement by multi-focus imaging and is widely applied in industrial inspection and biomedical research. However, the inherent trade-off between depth of field and lateral resolution of the microscope lens results in EDOF's axial resolution being significantly lower than its lateral resolution, limiting its axial recognition sensitivity. To address this, we propose a Kalman filter based spatial-compensated 3D microscopy measurement method. It introduces the multi-view measurement strategy and capitalizes on the anisotropic resolution characteristics of the single-view EDOF. By employing a Kalman filter to fuse multi-view measurement data, the proposed method achieves compensation from higher lateral resolution to lower axial. While expanding the spatial resolution field-of-view, we significantly enhance axial resolution. Experimental results demonstrate that, compared to conventional EDOF, the proposed method improves the axial resolution of the original single-view by approximately 75%. It effectively reduces the anisotropy of resolution and offers a novel, to the best of our knowledge, approach for more comprehensively characterizing the 3D morphology of complex surfaces.