運用數位微鏡裝置發展全域性掃描準單擊式共軛焦顯微形貌量測技術與系統

Development of quasi One-shot Full-field Confocal Microscopic Surface Profilometry and its Measuring Systems using Digital Micro-mirror Device

阮忠德(Duc-Trung Nguyen)

指導教授 ： 陳亮嘉

國立臺灣大學/工學院/機械工程學研究所/碩士(2016年)

https://doi.org/10.6342/NTU201601083

摘要

本論文中，運用數位微鏡裝置(DMD)的準單擊式全域性掃描共軛焦顯微形貌量測術與量測系統獲得提出與執行。藉由精準的控制數位微鏡裝置上每一片微米鏡片來排列結構圖形，此微鏡裝置可視為針孔陣列，用以達到全域式共軛焦量測，不用任何橫向機械式掃描，同時能避免傳統共焦顯微術中遭遇到訊號橫向交談問題。研究中被發展出的兩個創新方法完全省略共軛焦垂直掃描。
第一個方法中，一個新方法被發展用來編碼並解碼表面形貌的深度資訊，藉由量測深度和相對應的共軛焦顯微鏡架構中表面繞射圖案之間的相關性。此發展系統可以將量測點與其周圍的光橫向交談最小化。因此，此系統在表面重建上可以有更高的強健性與效能和較少的橫向掃描可以說是準單擊式量測。
第二種方法利用彩色共焦顯微機構與寬頻光源，在深度-色散物鏡的量測範圍內，從偵測的光譜響應中編碼與解碼深度資訊。一個發展的訊號處理演算法可以建立一條光譜響應曲線，基於多光譜訊號擷取自擁有多頻光譜偵測功能的IMEC面形光譜相機。表面深度藉由偵測擁有最大光強的波長可以被精準的決定偵測光譜響應曲線中的峰值以取得最大光強值對應的波長。更重要的是，運用DMD排列設計的虛擬針孔陣列，將可使共軛焦橫向交談影響最小化。
根據使用二維正規化相關係數來建立正規化相關係數-深度響應曲線、在基於DMD的數位的繞射-共軛焦成像相關性原理、以及包含面型光譜相機的彩色共軛焦系統中運用一維正規化相關係數在光譜響應匹對程序，以上四項技術的發展，以克服光源能量波動或待測面反射率不同所造成的光強變化問題。
為檢驗被提出的方法之可行性與驗證系統的量測精度，用所發展的量測系統來測量校正過並擁有鏡面反射的階高塊。經由來自鏡射表面樣品的初步實驗結果，已證實基於DMD的數位的繞射-共軛焦成像相關性顯微鏡，在400 μm的量測範圍內高度量測的重複性在一個標準差是0.03 μm，其量測的最大誤差為0.1％在可量測深度範圍內。另一方面，包含面型光譜相機的彩色共軛焦系統中，用鏡面型樣品的隨機標準差低於0.12 μm。將系統用在樣品表面粗糙度為Ra=1.6的案例中，高度量測的一個標準差可以保持於0.67 μm以內。

關鍵字

自動化光學檢測 ； 彩色共軛焦顯微術 ； 表面輪廓量測 ； 正規化相關係數 ； 全域量測 ； 數位微鏡裝置(DMD)

並列摘要

In the thesis, quasi one-shot full-field confocal microscopic surface profilometry and its measuring systems using digital micro-mirror device (DMD) is proposed and implemented. With precisely controlling each micro-mirror on DMD and modulated with structured patterns, a DMD can function as an array of pinholes to perform full-field confocal measurement without any lateral mechanical scanning and to avoid suffering lateral signal cross talk encountered in conventional confocal microscopy. Two novel approaches were developed in the study to omit confocal vertical scanning fully.
In the first approach, a new method was developed to encode and decode depth information of surface profile by the correlation between a measuring depth and its corresponding surface diffraction image pattern in confocal microscopic configuration. The developed system can minimize the light cross talk between each measuring point and its neighboring region. Thus, it can be more robust and efficient in surface reconstruction with less lateral scanning in a manner of quasi-one shot measurement.
The second approach uses a chromatic confocal microscope setup with broadband light, in which a depth-dispersion chromatic objective is used to encode and decode depth information from the detected spectrum response receiving at its measuring depth range. A developed signal processing algorithm can establish a spectrum response curve based on the multi-band spectral information captured by using an IMEC area spectral camera with multi-band spectrum detection capability. The surface depth can be accurately determined by detecting the wavelength with maximum intensity respected to its detected surface depth. More importantly, undesired confocal cross talk effect can be also minimized by applying a designed virtual pinhole array formed by DMD.
By applying 2-D normalized cross correlation to establish normalized cross correlation–depth response curve in the developed digital diffractive-confocal imaging correlation microscope, which is based on DMD and 1-D normalized cross correlation for spectrum response matching procedure in chromatic confocal system including spectral area camera, the developed confocal microscopes can be immune to light intensity variation being caused by light source’s power fluctuating and reflectivity variation of sample’s surface.
To test the feasibility of the proposed method and verify its measurement accuracy, a pre-calibrated step height with specular reflection was measured by the developed measuring systems. From the preliminary experimental results with specular surface sample, it was verified that for digital diffractive-confocal imaging microscope based on DMD, the repeatability with one-standard deviation on height measurement is 0.03 μm in a measuring rage of 400 μm while the maximum measured error is 0.1 % of the measurable range. On the other hand, for chromatic confocal system including spectral area camera, one standard deviation is lower than 0.12 μm for mirror-like sample. Applying the system in the case of surface sample with roughness Ra = 1.6, one-standard deviation on height measurement still can be kept lower than 0.67 μm.

並列關鍵字

Automated optical inspections (AOI) ； chromatic confocal microscopy ； surface profilometry ； normalized cross correlation (NCC) ； full-field measurement ； digital micro-mirror device (DMD)

參考文獻

1. T. R. Corle and G. S. Kino, "CHAPTER 1 - Introduction," in Confocal Scanning Optical Microscopy and Related Imaging Systems (Academic Press, Burlington, 1996), pp. 1-66.

2. "Bibliography on confocal microscopy and its applications," Scanning 16, 33-56 (1994).

3. T. Wilson and B. R. Masters, "Confocal microscopy," Applied Optics 33, 565-566 (1994).

4. C. J. R. Sheppard and X. Q. Mao, "Confocal Microscopes with Slit Apertures," Journal of Modern Optics 35, 1169-1185 (1988).

5. T. Tanaami, S. Otsuki, N. Tomosada, Y. Kosugi, M. Shimizu, and H. Ishida, "High-speed 1-frame/ms scanning confocal microscope with a microlens and Nipkow disks," Applied Optics 41, 4704-4708 (2002).