Meta-Optics for Image Processing and Object Classification

Image processing has become a critical technology in a variety of science and engineering disciplines. While most image processing is performed digitally, optical analog processing has the advantages of being low-power and high-speed though it requires a large volume. Meta-optics provide the advantage of thin form factor optics while also allowing complex transfer functions to be employed. In this talk, I will discuss the use of meta-optics for applications in image processing as well as object classification. Specifically, I will discuss the use of meta-optics as optical front-ends that perform edge filtering as well as identification of higher level spatial features for object classifiers. The meta-optics are designed using end-to-end optimization so that the optical front-end and digital back-end work in harmony. This architecture allows for multiple metasurfaces to be optimized while also allowing us to incorporate noise in the design loop, resulting in robust experimental systems. The meta-optics processors are designed for both coherent and incoherent illumination, enabling a wide range of use cases that take advantage of both a reduction in processing time and power consumption.