Assessing Noise Impact on DeepOrientation - A Convolutional Neural Network for Local Fringe Orientation Map Estimation

Abstract

This paper discusses noise robustness of DeepOrientation, a convolutional neural network developed for fast and accurate local fringe orientation map estimation, enhancing full-field optical measurement techniques such as interferometry and holographic microscopy. The use of neural networks to determine the final result of the optical measurement may raise legitimate metrological concerns and therefore we still recommend the use of fully mathematically sound solutions for both fringe pattern prefiltration and phase retrieval. DeepOrientation does not replace mathematically rigorous algorithms but supports them providing fringe orientation map vital for 2D Hilbert transform phase demodulation, requiring prefiltered fringe data for optimal performance. We analyze DeepOrientation sensitivity to prefiltration noise-related accuracy using simulated data and validate results with experimentally recorded fringe patterns.

Full Text: PDF

References

1. G. Popescu, Quantitative Phase Imaging of Cells and Tissues, (McGraw Hill Professional, 2011). DirectLink
2. D. Malacara, M. Servin, Z. Malacara, Interferogram Analysis for Optical Testing (Marcel Dekker, 1998). DirectLink
3. M. Takeda, H. Ina, S. Kobayashi, "Fourier-transform method of fringe-pattern analysis for computer-based topography and interferometry", J. Opt. Soc. Am. 72(1), 156 (1982). CrossRef
4. Q. Kemao, "Windowed Fourier transform for fringe pattern analysis", Appl. Opt. 43(13), 2695 (2004). CrossRef
5. K.G. Larkin, D.J. Bone, M.A. Oldfield, "Natural demodulation of two-dimensional fringe patterns. I. General background of the spiral phase quadrature transform", J. Opt. Soc. Am. A 18(8), 1862 (2001). CrossRef
6. M. Trusiak, M. Cywinska, V. Mico, J.-A. Picazo-Bueno, C. Zuo, P. Zdankowski, K. Patorski, "Variational Hilbert Quantitative Phase Imaging", Sci Rep 10, 13955 (2020). CrossRef
7. B. Jahne, Practical Handbook on Image Processing for Scientific Applications (CRC, 1997). CrossRef
8. F. Zhang, W. Liu, J. Wang, Y. Zhu, and L. Xia, "Anisotropic partial differential equation noise-reduction algorithm based on fringe feature for ESPI", Opt. Commun. 282(12), 2318 (2009). CrossRef
9. J. Villa, J. A. Quiroga, and I. De la Rosa, "Regularized quadratic cost function for oriented fringe-pattern filtering", Opt. Lett. 34(11), 1741 (2009). CrossRef
10. Q. Yu, X. Liu, and X. Sun, "Generalized spin filtering and an improved derivative-sign binary image method for the extraction of fringe skeletons", Appl. Opt. 37(20), 4504 (1998). CrossRef
11. H. Wang, Q. Kemao, "Frequency guided methods for demodulation of a single fringe pattern", Opt. Express 17(17), 15118 (2009). CrossRef
12. J.L. Marroquin, R. Rodriguez-Vera, M. Servin, "Local phase from local orientation by solution of a sequence of linear systems", J. Opt. Soc. Am. A 15(6), 1536 (1998). CrossRef
13. M. Cywińska, M. Rogalski, F. Brzeski, K. Patorski, M. Trusiak, "DeepOrientation: convolutional neural network for fringe pattern orientation map estimation", Opt. Express 30(23), 42283 (2022). CrossRef
14. M. Cywińska, M. Trusiak, K. Patorski, "Automatized fringe pattern preprocessing using unsupervised variational image decomposition", Opt. Express 27(16), 22542 (2019). CrossRef
15. X. Yang, Q. Yu, S. Fu, "A combined method for obtaining fringe orientations of ESPI", Opt. Commun. 273(1), 60 (2007). CrossRef