High-speed optical measurement systems are wholly constrained by the number of measurements that can be acquired in a limited amount of time. Unfortunately, most of these painstakingly acquired measurements are wasted collecting much more data than is required to accurately determine a given signal of interest. Specifically, real-world signals (e.g. images) are highly compressible and can be accurately represented by relatively few significant coefficients in an appropriate mathematical basis. Traditionally a signal is sampled in the physical domain according to the Nyquist theorem to acquire a raw digital representation and then a compression algorithm is applied, which eliminates as much of the redundancy in the original data as possible. Hence, most of the acquired data is essentially thrown away and, consequently, for most applications in high-speed measurement the raw data bandwidth is far larger than is truly necessary. In this talk, we will discuss our recent research in applying optical signal processing and compressed sensing to enhance performance in such high-speed measurement-limited applications. Compressed sensing is a recent and influential sampling paradigm that advocates a more efficient signal acquisition process by implementing image compression directly in the physical layer. Specifically, we will discuss our research into constructing compressed sensing based optical hardware systems for high-throughput microscopy, optical coherence tomography, LIDAR, and hyperspectral imaging.
|