Off-Focus-Image-Restoration-Based Three-Dimensional Particle Localization for Improved Measurement Resolution

Ultrahigh-precision 3-D particle localization and tracking in microscopic vision systems plays an important role in numerous micro/nanoscale applications. However, it remains a challenge due to image degradation caused by lens defocusing. Here, we propose an off-focus-image-restoration-based approach to enhance measurement resolution in 3-D particle localization. In this method, a point spread function corresponding to the off-focus position is established and served as the deconvolution kernel. By employing an improved nonlinear filtering algorithm, high-quality restored particle images are reconstructed effectively. The principle of the proposed method is presented in detail, followed by experimental validation in particle localization and 3-D trajectory tracking. With the proposed approach, results demonstrate that measurement resolution along the lateral and axial directions is improved by 38% and 31%, respectively. The proposed method provides a practical solution for 3-D particle tracking within a relatively large axial range, which is believed to be significant for micro/nanoscale spatial-localization-related applications.