Auto-alignment for the Focusing Unit of NFR system by Pattern recognition of Neural Network

摘要

In this paper, we propose the auto-alignment methodology for the focusing unit of the NFR system using the pattern recognition by the neural network. By the inference fringe pattern analysis, proper features were selected and the neural network was trained. Using the trained neural network, we verified the proposed methodology. By the five axis feedback position control, we carried out the auto-alignment of the FU and obtained the improved aligned state from a random state. The process was error decreasing process and the RMS wavefront also was I improved.