Chinese Optics, Volume. 16, Issue 3, 637(2023)
White light interferometry micro measurement algorithm based on principal component analysis
Fig. 1. Schematic diagram of white light microscopic interferometry
Fig. 2. Relationship between white light interference intensity and optical path difference
Fig. 3. (a) The first and (b) the second white light phase shifting interferograms
Fig. 4. Comparison of calculation results of ideal height of step and real height of step
Fig. 5. White light phase-shifting interferogram of step (The red line represents an actual physical length of 0.2 mm)
Fig. 6. The heights of step caculated by (a) PCA-WLI algorithm and (b) AIA algorithm
Table 1. Comparison of PCA-WLI and AIA algorithms
View tableView in ArticleTable 1. Comparison of PCA-WLI and AIA algorithms
算法 PV误差（%） RMS误差（%） 计算时间（s） PCA-WLI算法 0.452 0.035 0.026 AIA算法 0.443 0.034 5.2465
Table 2. Measurement results of step by different algorithms
View tableView in ArticleTable 2. Measurement results of step by different algorithms
Algorithm Times PCA-WLI AIA 1 958.35 955.09 2 949.59 948.38 3 947.80 960.16 4 955.29 959.11 5 957.46 960.65 6 949.70 958.31 7 951.09 944.28 8 954.79 956.26 9 956.99 949.77 10 955.56 956.58 Mean value/nm 953.66 954.87
Table 3. Measurement results of surface roughness specimen
View tableView in ArticleTable 3. Measurement results of surface roughness specimen
Times Ra 1 24.9 2 25.3 3 24.7 4 24.6 5 25.1 6 24.8 7 24.2 8 24.6 9 25.5 10 24.6 Mean value/nm 24.83 Standard deviation/nm 0.3831
Table 4. Comparison of consuming times of interferogram sequence with different resolutions processed by the two algorithms (s)
View tableView in ArticleTable 4. Comparison of consuming times of interferogram sequence with different resolutions processed by the two algorithms (s)
算法 分辨率 200×200 400×400 800×800 PCA-WLI算法 0.0168 0.0358 0.1226 AIA算法 2.8852 10.0738 38.8263
Get Citation
Copy Citation Text
Hao-bo CHEN, Li-wei ZHANG, Wen-qing SUN, Bao-hua CHEN, Zhao-liang CAO, Quan-ying WU. White light interferometry micro measurement algorithm based on principal component analysis[J]. Chinese Optics, 2023, 16(3): 637
Category: Original Article
Received: Jul. 26, 2022
Accepted: Sep. 9, 2022
Published Online: May. 31, 2023
The Author Email:
