Investigation of unsupervised and supervised hyperspectral anomaly detection

Hyperspectral sensing is a valuable tool for detecting anomalies and distinguishing between materials in a scene. Hyperspectral anomaly detection (HS-AD) helps characterize the captured scenes and separates them into anomaly and background classes. It is vital in agriculture, environment, and military applications such as RSTA (reconnaissance, surveillance, and target acquisition) missions. We previously designed an equal voting ensemble of hyperspectral unmixing and three unsupervised HS-AD algorithms. We later utilized a supervised classifier to determine the weights of a voting ensemble, creating a hybrid of heterogeneous unsupervised HS-AD algorithms with a supervised classifier in a model stacking, which improved detection accuracy. However, supervised classification methods usually fail to detect novel or unknown patterns that substantially deviate from those seen previously. In this work, we evaluate our technique and other supervised and unsupervised methods using general hyperspectral data to provide new insights.

Impact of regularization on achieved resolution in 3D tunable structured illumination microscopy (TSIM)

We present a study that evaluates the impact of regularization on the achieved resolution in restorations from a novel three-dimensional (3D) Structured Illumination Microscopy (3D-SIM) system with desirable tunability properties. This contribution is the first performance evaluation of the Tunable SIM (TSIM) system through the restoration process. The study quantifies the achieved resolution in restorations, from simulated TSIM data of a 3D star-like object, at various expected resolution limits controlled by system parameters, and at different noise levels mitigated by the Generalized Wiener filter, a computationally efficient method, successfully applied to other conventional 3D-SIM systems. We show that theoretical TSIM resolution limits are attained in the absence of noise, while with increasing noise levels, the necessary increase in regularization and residual restoration artifacts contributed to a $\sim$ 5%-10% and a 20% reduction in the axial achieved resolution, in 20-dB and 15-dB data, respectively, which is within the pixel size (20 nm) limitation.