超快雷射又稱超短脈衝雷射(ultrashort pulse laser)，將能量集中在皮秒或飛秒等級以內打出，以激發出生物或材料的非線性光學性質，因此在多光子顯微技術、超快雷射加工等應用上已成為不可或缺的要角，然而由於本身雷射在時域上飛秒等級的脈衝寬度，無法直接使用示波器觀測，通常需要用干涉架構、倍頻訊號與光譜資訊等間接手法來還原，使超快雷射在校正技術上困難重重。目前已成熟的超快雷射量測技術有，可還原時域脈衝寬度的光學自相關干涉儀(optical autocorrelator)、可重建雷射電場的頻域光學解析開關(frequency resolved optical gating，FROG)，以及可量測雷射頻域相位的空間頻率干涉數(spatial spectral interferometry，SSI)。本研究引入深度學習技術，針對SSI建立一套新的深度學習加速模型，以取代傳統費時的還原演算法。憑藉深度學習在影像處理上的成功，本研究基於卷積神經網路(convolution neural network，CNN)建立四層卷積層搭配平均池化層(average pooling)與兩層全連接層(fully connected layer，FC)之模型，並以SSI影像作為輸入；頻域相位作為模型輸出，訓練後的模型不僅成功精準的還原雷射頻域相位，並在速度上也較傳統演算法有2.2倍的提升。同時在光學系統上，本研究也基於殘差神經網路(residual networks，ResNet)找出關鍵元件(linear deformable mirror，LDM)個別通道電壓與頻域相位的直接關係，訓練後的模型有能力從量測到的相位直接還原系統上LDM各個通道的電壓，取代了過往需要大量迭代與假設的鑑別方法。最後本研究嘗試將神經網路模型運行在不同平台與IDE (integrated development environment)上，比較各別的推論速度，為後續的系統整合提供可能的解決方案。

There are several well-established measurement techniques for ultrafast lasers. These include optical autocorrelators, which can restore the temporal pulse width; frequency-resolved optical gating (FROG) and spatial spectral interferometry (SSI), which can measure the laser spectral phase.
In this study, deep learning techniques are introduced to establish a new deep learning acceleration model for SSI, aimed at replacing traditional time-consuming reconstruction algorithms. Building upon the success of deep learning in image processing, this research utilizes a four-layer convolutional layers with average pooling and two fully connected layers (FC) to process SSI images as input and predict spectral phase as the model output. The trained model not only successfully and accurately restores the laser spectral phase but also demonstrates a 2.2-fold improvement in speed compared to traditional algorithms. Additionally, in the optical system, this research employs deep residual learning (ResNet) to identify the direct relationship between voltage combination of the linear deformable mirror (LDM) and spectral phase. The trained model has the capability to directly restore the voltage of each LDM channel from the measured phase, eliminating the need for extensive iterations and assumption-based identification methods used in the past. Finally, this study explores the implementation of the neural network model on different platforms and IDEs, comparing individual analysis speed and providing potential solutions for future system integration.

1. M. Göppert‐Mayer, “Elementary processes with two quantum transitions,” Ann. Phys. 18(78), 466479 (2009).
2. W. Denk, J. H. Strickler, and W. W. Webb, “Two-Photon Laser Scanning Fluorescence Microscopy,” Science. 248(4951), 7376 (1990).
3. W. R. Zipfel, R. M. Williams, and W. W. Webb,” Nonlinear magic: multiphoton microscopy in the biosciences,” Nat Biotechnol. 21(11), 13691377 (2003).
4. P. J T. Wang and C. Xu, “Three-photon neuronal imaging in deep mouse brain,” Optica 7(8), 947960 (2020).
5. S.V. Plotnikov, A.M. Kenny, and S.J. Walsh, “Measurement of muscle disease by quantitative second-harmonic generation imaging,” J Biomed Opt. 13(4), 044018 (2008).
6. L. Perrin, B. Bayarmagnai, and B. Gligorijevic, “Frontiers in intravital multiphoton microscopy of cancer,” Cancer Rep. 3(1), e1192 (2020).
7. R. R. Gattass, and E. Mazur, “Femtosecond Laser Micromachining in Transparent Materials,” Nat. Photon. 2, 219–225 (2008).
8. G. Bonamis, E. Audouard, C. Hönninger, J. Lopez, K. Mishchik, E. Mottay, and I. Manek-Hönninger, “Systematic study of laser ablation with GHz bursts of femtosecond pulses,” Opt. Express 28(19), 27702–27714 (2020).
9. M. Malinauskas, A. Žukauskas, S. Hasegawa, “Ultrafast laser processing of materials: from science to industry,” Light Sci Appl 5, e16133 (2016).
10. H Lin, C Wei, G Wang, H Chen, L Lin, M Ni, J Chen, S Zhuo, “Automated classification of hepatocellular carcinoma differentiation using multiphoton microscopy and deep learning,” J. Biophoton., 12(7), 2019.
11. Mikko J. Huttunen, Radu Hristu, Adrian Dumitru, Iustin Floroiu, Mariana Costache, and Stefan G. Stanciu, "Multiphoton microscopy of the dermoepidermal junction and automated identification of dysplastic tissues with deep learning," Biomed. Opt. Express 11(1), 186-199 (2020)
12. H. P. Weber, “Method for Pulsewidth Measurement of Ultrashort Light Pulses Generated by Phase-Locked Lasers using Nonlinear Optics,” J. Appl. Phys. 38(5), 22312234 (1967)
13. J. A. Armstrong, “MEASUREMENT OF PICOSECOND LASER PULSE WIDTHS,” Appl. Phys. Lett. 10(1), 1618 (1967).
14. J-C. M. Diels, J. J. Fontaine, I. C. McMichael, and F. Simoni, “Control and measurement of ultrashort pulse shapes (in amplitude and phase) with femtosecond accuracy,” Appl. Opt. 24(9), 12701282 (1985).
15. R. Trebino and D. J. Kane, “Using phase retrieval to measure the intensity and phase of ultrashort pulses: frequency-resolved optical gating,” J. Opt. Soc. Am. A 10(5), 11011111 (1993).
16. R. Trebino, Frequency-Resolved Optical Gating: The Measurement of Ultrashort Laser Pulses (Kluwer Academic, 2002).
17. P. O'Shea, S. Akturk, M. Kimmel, and R. Trebino, “Practical issues in ultra-short-pulse measurements with 'GRENOUILLE',” Appl. Phys. B 79, 683-691 (2004).
18. M. A. Krumbügel, C. L. Ladera, K. W. DeLong, D. N. Fittinghoff, J. N. Sweetser, and R. Trebino, “Direct ultrashort-pulse intensity and phase retrieval by frequency-resolved optical gating and a computational neural network,” Opt. Lett. 21(2), 143–145 (1996).
19. R. Ziv, A. Dikopoltsev, T. Zahavy, I. Rubinstein, P. Sidorenko, O. Cohen, and M. Segev, “Deep learning reconstruction of ultrashort pulses from 2D spatial intensity patterns recorded by an all-in-line system in a single-shot,” Opt. Express 28(5), 7528–7538 (2020).
20. T. Zahavy, A. Dikopoltsev, D. Moss, G. I. Haham, O. Cohen, S. Mannor, and M. Segev, “Deep learning reconstruction of ultrashort pulses,” Optica 5(5), 666–673 (2018).
21. S. Kleinert, A. Tajalli, T. Nagy, and U. Morgner, “Rapid phase retrieval of ultrashort pulses from dispersion scan traces using deep neural networks,” Opt. Lett. 44(4), 979–982 (2019).
22. P. V. Kolesnichenko and D. Zigmantas, “Neural-network-powered pulse reconstruction from one-dimensional interferometric correlation traces,” Opt. Express 31(7), 11806–11819 (2023)
23. D. Meshulach, D. Yelin, and Y. Silberberg, “Real-time spatial-spectral interference measurements of ultrashort optical pulses,” J. Opt. Soc. Am. B 14, 2095–2098 (1997).
24. P. Bowlan, P. Gabolde, A. Schreenath, K. McGresham, and R. Trebino, “Crossed-beam spectral interferometry: a simple, high-spectral-resolution method for completely characterizing complex ultrashort pulses in real time,” Opt. Express 14, 11892–11900 (2006).
25. P. Bowlan, P. Gabolde, M. A. Coughlan, R. Trebino, and R. J. Levis, “Measuring the spatiotemporal electric field of ultrashort pulses with high spatial and spectral resolution,” J. Opt. Soc. Am. B 25(6), A81A91 (2008).
26. D. N. Fittinghoff, J. L. Bowie, J. N. Sweetser, R. T. Jennings, M. A. Krumbügel, K. W. DeLong, R. Trebino, and I. A. Walmsley, “Measurement of the intensity and phase of ultraweak, ultrashort laser pulses,” Opt. Lett. 21, 884–886 (1996).
27. Swamp Optics, “Autocorrelation,” swampoptics.com, 2019. [Online]. Available: http://www.swampoptics.com/autocorrelation.html. [Accessed Aug. 12, 2021].
28. H. Folliot, M. Lynch, A. L. Bradley, T. Krug, L. A. Dunbar, J. F. Donegan, and L. P. Barry, “Two-photon-induced photoconductivity enhancement in semiconductor microcavities: A theoretical investigation,” J. Opt. Soc. Amer. B, 19(10), 2396-2402 (2002).
29. M. Sheik-Bahae, D. C. Hutchings, D. J. Hagan and E. W. Van Stryland, ”Dispersion of bound electron nonlinear refraction in solids,” IEEE Journal of Quantum Electronics, 27(6), 1296-1309 (1991).
30. E. Ochoa-Martínez, L. Barrutia, M. Ochoa, E. Barrigón, I. García, I. Rey-Stolle, C. Algora, P. Basa, G. Kronome, M. Gabás, “Refractive indexes and extinction coefficients of n- and p-type doped GaInP, AlInP and AlGaInP for multijunction solar cells,” Solar Energy Materials and Solar Cells, 174, 388-396 (2018).
31. D. Meshulach, Y. Barad, and Y. Silberberg, “Measurement of ultrashort optical pulses by third-harmonic generation,” J. Opt. Soc. Am. B 14, 2122-2125 (1997).
32. S. Benis, C. M. Cirloganu, N. Cox, T. Ensley, H. Hu, G. Nootz, P. D. Olszak, L. A. Padilha, D. Peceli, M. Reichert, S. Webster, M. Woodall, D. J. Hagan, and E. W. Van Stryland, “Three-photon absorption spectra and bandgap scaling in direct-gap semiconductors,” Optica 7, 888-899 (2020).
33. W. Zhao, W. Jiang & X. Qiu, “Deep learning for COVID-19 detection based on CT images,” Sci Rep 11, 14353 (2021).
34. K.A. Tran, O. Kondrashova, A. Bradley, “Deep learning in cancer diagnosis, prognosis and treatment selection,” Genome Med. 13(1), 152 (2021).
35. A.J. Lew, CH. Yu, YC. Hsu, “Deep learning model to predict fracture mechanisms of graphene,” npj 2D Mater 5, 48 (2021).
36. B.Y. Tseng, C.W. Guo, Y.C. Chien, J.P. Wang, C.H. Yu, “Deep Learning Model to Predict Ice Crystal Growth,” Adv. Sci. 10(21), 2207731 (2023).
37. Krizhevsky, A., Sutskever, I., Hinton, G.: ImageNet classification with deep convolutional neural networks. In: NeurIPS (2012).
38. He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016).
39. Principle and framework of Residual Networks (ResNet), from:
https://d2l.ai/chapter_convolutional-modern/resnet.html
40. NVIDIA, specification of Jetson Nano developer kit, from:
https://developer.nvidia.com/embedded/jetson-nano-developer-kit
41. National Instrument, specification of sbRIO-9627, from:
https://www.ni.com/zh-tw/support/model.sbrio-9627.html
42. National Instrument, specification of cRIO-9039, from:
https://www.ni.com/zh-tw/support/model.crio-9039.html
43. Ngene, introduction of Deep Learning Toolkit, from:
https://www.ngene.co/deep-learning-toolkit-for-labview
44. Xilinx white paper, WP529 (v1.0) (2021)
45. Ngene, introduction of FPGA Add-on, from:
https://www.ngene.co/deepltk-accelerator-for-fpga
46. Coherent, specification of Chameleon Ultra series, from:
https://www.coherent.com/lasers/oscillators/chameleon-ultra
47. Servo Motor, specification of KDC101, from:
https://www.thorlabs.com/thorproduct.cfm?partnumber=KDC101
48. Daniel M. Mittleman. “14. Measuring Ultrashort Laser Pulses I: Autocorrelation”. University Lecture. 2015.
49. W. Rawat and Z. Wang, “Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review,” in Neural Computation, 29(9), 2352-2449 (2017).
50. Convolution operation diagram, from:
https://medium.com/@bdhuma/6-basic-things-to-know-about-convolution-daef5e1bc411
51. R.D. Rakshit, D.R. Kisku, P. Gupta, & J.K. Sing, “Cross-resolution face identification using deep-convolutional neural network,” Multimedia Tools and Applications, 80(14), 20733 - 20758 (2021).
52. Activation function table, from:
https://medium.com/@shrutijadon/survey-on-activation-functions-for-deep-learning-9689331ba092