本研究針對我們製作的三款原理與結構特性各異的紅外超穎透鏡:M1、M5與M6設計統一的量測系統檢測其光學性能。M1 與 M5 採用傳播相位(Propagation phase)設計，分別為焦距 2.36 mm、視角為 90° 的小尺寸短焦透鏡，以及焦距約 15.4 mm、視角為 60° 的大尺寸長焦透鏡。M6 為雙透鏡系統，採用 PB phase 設計，由兩顆直徑 1.65 mm 的圓形透鏡並排而成，其焦距為2 mm 能分別聚焦左旋與右旋圓偏振紅外光。三款透鏡皆設計應用於 8 – 12 μm 波段熱影像系統中。 我們使用雷射切割製作不同線寬的光柵圖形直接觀察三者的影像品質，並量測其調製傳遞函數（modulation transfer function, MTF），量化其光學性能的表現，並觀察其在不同工作距離下的影像變化。此外，我們也說明了量測光學性能時，須注意哪些架構上、或是影像處理上的問題。包括測量各個視角性能的方法；或是我們針對成像系統內部或測試物表面反射造成的雜訊設計實驗，對其做抗反射處理，以提升量測準確性；以及如何訂定適當的物距，才能得到截止空間頻率之下的MTF值。雖然結果顯示量測數據與模擬結果存在差異，我們仍從多次實驗中累積了寶貴經驗，並發現許多在量測透鏡光學性能時應特別注意的細節，這將有助於未來進一步優化整體實驗流程與設計準確性。 ;This study evaluates the optical performance of three infrared metalenses—M1, M5, and M6—each designed with distinct principles and structural characteristics, using a unified measurement system. Both M1 and M5 adopt a propagation phase design. M1 is a compact short-focus lens with a focal length of 2.36 mm and a 90° field of view (FOV), while M5 is a larger long-focus lens with a focal length of approximately 15.4 mm and a 60° FOV. M6 is a dual-lens system based on the Pancharatnam–Berry (PB) phase design, consisting of two 1.65 mm diameter circular lenses placed side by side. It has a focal length of 2 mm and is capable of independently focusing left- and right-handed circularly polarized infrared light. All three lenses are designed for use in thermal imaging systems operating in the 8 – 12 μm wavelength range. We fabricated grating targets with varying linewidths using laser cutting to directly observe the image quality of each lens. The modulation transfer function (MTF) was measured to quantitatively assess optical performance and evaluate image variations at different working distances. Additionally, we addressed critical considerations in system architecture and image processing when measuring optical characteristics. These include methods for evaluating performance across different viewing angles, experimental designs to suppress reflection-induced noise from the imaging system or test object surfaces through anti-reflection treatments, and strategies for selecting appropriate object distances to obtain MTF values at the cutoff spatial frequency. Although the experimental results showed noticeable discrepancies from the simulations, we gained valuable insights from repeated experiments. These findings revealed critical details that must be carefully considered when measuring the optical performance of the lens, and contribute to improving future measurement strategies.
