在當今世界中，光學透鏡在我們的日常生活中佔據的很重要的地位，其廣泛的被運用在各種科技設備中，包括智慧型手機、自動駕駛傳感器和擴增實境(Augmented Reality, AR)及虛擬實境(Virtual Reality, VR)設備中，這些設備都有朝著更薄、更輕的消費電子產品的趨勢，創造了對光學元件不斷微型化的需求。然而，傳統的折射透鏡，如凸透鏡或凹透鏡材料，因為其體積龐大，嚴重阻礙了光學元件的微型化。而近年來，在次波長尺度上的奈米結構發展取得了顯著進展，對超穎表面的關注日益增加。這些奈米結構超穎表面利用集體共振來在奈米尺度上控制電磁波的特性，這一進展為創建微型光學元件展開了可能性。我們對超穎透鏡進行了光學特性的分析，並利用兩個連續的人工智能模型來解決超穎透鏡拍攝的影像中由於超穎透鏡的材料損耗和結構散射所導致的模糊與色偏問題。在人工智能模型這部分，我們分別使用自動編碼器和CodeFormer來校正色偏和重建影像細節。我們透過自動編碼器模型成功解決了所有面部影像的色偏，而CodeFormer模型則可以有效地重建了標準正面臉部、帶有臉部表情的面部細節和側面臉部影像，透過這樣的連續兩個人工智慧模型提升了超穎透鏡在日常生活的應用潛力。

In today’s world, optical lenses play a vital role in our daily lives and are widely used in various technological devices, including smartphones, self-driving sensors, and augmented reality (AR) / virtual reality (VR) equipment. These devices are trending towards thinner and lighter consumer electronics, creating a demand for the continuous miniaturization of optical components. However, traditional refractive lenses, such as convex or concave materials, are bulky and severely hinder the miniaturization of optical components.
In recent years, there has been significant progress in the development of nanostructures on sub-wavelength scales, leading to a growing interest in metasurfaces. These nanostructured metasurfaces utilize collective resonances to control the characteristics of electromagnetic waves at the nano-scale, opening up possibilities for the creation of miniature optical components.
We conducted an analysis of the optical properties of the metasurface lens and used two sequential AI models to address the blurriness and color cast issues in images captured by the metasurface lens due to material loss and structural scattering of the metasurface lens. In terms of AI models, we used an Autoencoder and CodeFormer to correct color cast and reconstruct image details, respectively. We successfully addressed color cast in all facial images using the Autoencoder model, while the CodeFormer model effectively reconstructed standard frontal faces, facial expressions, and side profile images. Through these two sequential AI models, we have increased the potential for the application of metasurface lenses in daily life.

[1] Lan, G., Wang, Y., & Ou, J. Y. (2022). Optimization of metamaterials and metamaterial-microcavity based on deep neural networks. Nanoscale Advances, 4(23), 5137-5143.
[2] Yang, J., Ghimire, I., Wu, P. C., Gurung, S., Arndt, C., Tsai, D. P., & Lee, H. W. H. (2019). Photonic crystal fiber metalens. Nanophotonics, 8(3), 443-449.
[3] Chen, M. K., Wu, Y., Feng, L., Fan, Q., Lu, M., Xu, T., & Tsai, D. P. (2021). Principles, functions, and applications of optical meta‐lens. Advanced Optical Materials, 9(4), 2001414.
[4] Hu, J., Bandyopadhyay, S., Liu, Y. H., & Shao, L. Y. (2021). A review on metasurface: from principle to smart metadevices. Frontiers in Physics, 8, 586087.
[5] Hsu, W. L., Chen, Y. C., Yeh, S. P., Zeng, Q. C., Huang, Y. W., & Wang, C. M. (2022). Review of metasurfaces and metadevices: advantages of different materials and fabrications. Nanomaterials, 12(12), 1973.
[6] Sun, S., Yang, K. Y., Wang, C. M., Juan, T. K., Chen, W. T., Liao, C. Y., ... & Tsai, D. P. (2012). High-efficiency broadband anomalous reflection by gradient meta-surfaces. Nano letters, 12(12), 6223-6229.
[7] Babicheva, V. E., & Evlyukhin, A. B. (2017). Resonant lattice Kerker effect in metasurfaces with electric and magnetic optical responses. Laser & Photonics Reviews, 11(6), 1700132.
[8] Falcone, F., Lopetegi, T., Laso, M. A. G., Baena, J. D., Bonache, J., Beruete, M., ... & Sorolla, M. (2004). Babinet principle applied to the design of metasurfaces and metamaterials. Physical review letters, 93(19), 197401.
[9] Yu, C. Y., Zeng, Q. C., Yu, C. J., Han, C. Y., & Wang, C. M. (2021). Scattering analysis and efficiency optimization of dielectric Pancharatnam–Berry-Phase metasurfaces. Nanomaterials, 11(3), 586.
[10] Chen, K., Feng, Y., Monticone, F., Zhao, J., Zhu, B., Jiang, T., ... & Qiu, C. W. (2017). A reconfigurable active Huygens' metalens. Advanced materials, 29(17), 1606422.
[11] Ren, J., Li, T., Fu, B., Wang, S., Wang, Z., & Zhu, S. (2021). Wavelength-dependent multifunctional metalens devices via genetic optimization. Optical Materials Express, 11(11), 3908-3916.
[12] Aieta, F., Genevet, P., Kats, M. A., Yu, N., Blanchard, R., Gaburro, Z., & Capasso, F. (2012). Aberration-free ultrathin flat lenses and axicons at telecom wavelengths based on plasmonic metasurfaces. Nano letters, 12(9), 4932-4936.
[13] Aieta, F., Kats, M. A., Genevet, P., & Capasso, F. (2015). Multiwavelength achromatic metasurfaces by dispersive phase compensation. Science, 347(6228), 1342-1345.
[14] Shrestha, S., Overvig, A. C., Lu, M., Stein, A., & Yu, N. (2018). Broadband achromatic dielectric metalenses. Light: Science & Applications, 7(1), 85.
[15] Wang, S., Wu, P. C., Su, V. C., Lai, Y. C., Chen, M. K., Kuo, H. Y., ... & Tsai, D. P. (2018). A broadband achromatic metalens in the visible. Nature nanotechnology, 13(3), 227-232.
[16] Fan, Z. B., Qiu, H. Y., Zhang, H. L., Pang, X. N., Zhou, L. D., Liu, L., ... & Dong, J. W. (2019). A broadband achromatic metalens array for integral imaging in the visible. Light: Science & Applications, 8(1), 67.
[17] Chen, W. T., Zhu, A. Y., Sanjeev, V., Khorasaninejad, M., Shi, Z., Lee, E., & Capasso, F. (2018). A broadband achromatic metalens for focusing and imaging in the visible. Nature nanotechnology, 13(3), 220-226.
[18] Chen, W. T., Zhu, A. Y., Sisler, J., Bharwani, Z., & Capasso, F. (2019). A broadband achromatic polarization-insensitive metalens consisting of anisotropic nanostructures. Nature communications, 10(1), 355.
[19] Fan, C. Y., & Su, G. D. J. (2021). Time-effective simulation methodology for broadband achromatic metalens using deep neural networks. Nanomaterials, 11(8), 1966.
[20] Wang, F., Geng, G., Wang, X., Li, J., Bai, Y., Li, J., ... & Zhou, J. (2022). Visible achromatic metalens design based on artificial neural network. Advanced Optical Materials, 10(3), 2101842.
[21] Kreyszig, E., Kreyszig, H., Norminton, E. J., & Corliss, S. (2011). Partial differential equations (PDEs). Advanced Engineering Mathematics, 10.
[22] Gaskill, J. D. (1978). Linear systems, Fourier transforms, and optics (Vol. 56). John Wiley & Sons.
[23] Pizer, S. M., Amburn, E. P., Austin, J. D., Cromartie, R., Geselowitz, A., Greer, T., ... & Zuiderveld, K. (1987). Adaptive histogram equalization and its variations. Computer vision, graphics, and image processing, 39(3), 355-368.
[24] Arici, T., Dikbas, S., & Altunbasak, Y. (2009). A histogram modification framework and its application for image contrast enhancement. IEEE Transactions on image processing, 18(9), 1921-1935.
[25] Krishna, A. S., Rao, G. S., & Sravya, M. (2013). Contrast enhancement techniques using histogram equalization methods on color images with poor lightning. International journal of computer science, engineering and applications, 3(4), 15.
[26] Loh, Y. P., & Chan, C. S. (2019). Getting to know low-light images with the exclusively dark dataset. Computer Vision and Image Understanding, 178, 30-42.
[27] Ledig, C., Theis, L., Huszár, F., Caballero, J., Cunningham, A., Acosta, A., ... & Shi, W. (2017). Photo-realistic single image super-resolution using a generative adversarial network. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4681-4690).
[28] Goodfellow, I. (2016). Nips 2016 tutorial: Generative adversarial networks. arXiv preprint arXiv:1701.00160.
[29] You, S., You, N., & Pan, M. (2019). PI-REC: Progressive image reconstruction network with edge and color domain. arXiv preprint arXiv:1903.10146.
[30] Zhang, Y., Li, K., Li, K., Wang, L., Zhong, B., & Fu, Y. (2018). Image super-resolution using very deep residual channel attention networks. In Proceedings of the European conference on computer vision (ECCV) (pp. 286-301).
[31] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).
[32] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
[33] Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. science, 313(5786), 504-507.
[34] Masci, J., Meier, U., Cireşan, D., & Schmidhuber, J. (2011). Stacked convolutional auto-encoders for hierarchical feature extraction. In Artificial Neural Networks and Machine Learning–ICANN 2011: 21st International Conference on Artificial Neural Networks, Espoo, Finland, June 14-17, 2011, Proceedings, Part I 21 (pp. 52-59). Springer Berlin Heidelberg.
[35] Zhou, S., Chan, K., Li, C., & Loy, C. C. (2022). Towards robust blind face restoration with codebook lookup transformer. Advances in Neural Information Processing Systems, 35, 30599-30611.
[36] Esser, P., Rombach, R., & Ommer, B. (2021). Taming transformers for high-resolution image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 12873-12883).
[37] Garipov, T., Izmailov, P., Podoprikhin, D., Vetrov, D. P., & Wilson, A. G. (2018). Loss surfaces, mode connectivity, and fast ensembling of dnns. Advances in neural information processing systems, 31.
[38] Choromanska, A., Henaff, M., Mathieu, M., Arous, G. B., & LeCun, Y. (2015, February). The loss surfaces of multilayer networks. In Artificial intelligence and statistics (pp. 192-204). PMLR.
[39] Taylor, D. P. (2002). Peak Signal-to-Noise Ratio in Digital Image Processing. In Encyclopedia of Computer Science and Technology (Vol. 43, No. 25, pp. 236-245). Taylor & Francis.
[40] Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image Quality Assessment: From Error Visibility to Structural Similarity. IEEE Transactions on Image Processing, 13(4), 600-612.
[41] Wikipedia contributors. (2023, August 15). Peak signal-to-noise ratio. Wikipedia. https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio