Modeling window smoothing effect hidden in fringe projection profilometry

Abstract

The reconstruction using MEMS micro-vibration mirrors is the most promising. However, the influence mechanism of the line laser, which serves as the sole light source of MEMS-based systems, on reconstruction is unclear. To fill this theoretical gap, we propose the first interpretable laser influence model called the window smoothing model. This model reveals the window smoothing effect of laser beam scanning and quantifies its impact on modulation intensity through a decay factor. Given the absence of lens defocusing issues, the window smoothing effect becomes the primary factor affecting the contrast of fringe patterns. Therefore, a novel method using laser parameters to evaluate system accuracy is introduced. Leveraging these insights, we further propose an optimal fringe number calibration method, also based on laser parameters, to maximize the reconstruction accuracy of any given system. Finally, experimental results validate the effectiveness of the proposed method. As the first model tailored for MEMS systems, the window smoothing model is interpretable, highly accurate, and universal. Therefore, this model can provide new insights and a theoretical foundation for achieving high-performance MEMS-based reconstruction.