VIS-NIR hyperspectral cameras

Hyperspectral and multispectral imaging can record a single scene across a range of spectral bands. The resulting three-dimensional dataset is called a "hypercube". A spectrum is available for each point of the image. This makes it possible to analyse, quantify or differentiate the elements and materials constituting the scene. This article presents the existing technologies on the market and their main characteristics in the VIS/NIR spectral domain (400-1000 nm). It then focuses on a specific multispectral technology called snapshot multispectral imaging, combining CMOS sensors and pixelated multispectral filters (filtering at the pixel level).

CHARACTERIZATION OF MODERN CCD AND CMOS SENSORS FOR SKY SURVEYS

Here we review the efforts we take in a newly established laboratory inside Institute of Physics in Prague in order to characterize modern large-format CCD and CMOS sensors for sky survey applications. While the laboratory is primarily established in order to participate in low-level CCD sensor characterization for LSST project, we also managed to perform a thorough laboratory testing of recently released Andor Marana sCMOS (which is especially interesting for wide-field sky monitoring applications due to its large format, backilluminated design, high achievable frame rate and low read-out noise), as well as detailed measurements of response non-linearity of Moravian Instruments G4-16000 CCD cameras (based on large-format Kodak KAF-16803 chip) used in several robotic telescopes. We briefly review the results acquired on these cameras, as well as hardware and software we developed for the laboratory.

Recognition of Cosmic Ray Images Obtained from CMOS Sensors Used in Mobile Phones by Approximation of Uncertain Class Assignment with Deep Convolutional Neural Network

In this paper, we describe the convolutional neural network (CNN)-based approach to the problems of categorization and artefact reduction of cosmic ray images obtained from CMOS sensors used in mobile phones. As artefacts, we understand all images that cannot be attributed to particles’ passage through sensor but rather result from the deficiencies of the registration procedure. The proposed deep neural network is composed of a pretrained CNN and neural-network-based approximator, which models the uncertainty of image class assignment. The network was trained using a transfer learning approach with a mean squared error loss function. We evaluated our approach on a data set containing 2350 images labelled by five judges. The most accurate results were obtained using the VGG16 CNN architecture; the recognition rate (RR) was 85.79% ± 2.24% with a mean squared error (MSE) of 0.03 ± 0.00. After applying the proposed threshold scheme to eliminate less probable class assignments, we obtained a RR of 96.95% ± 1.38% for a threshold of 0.9, which left about 62.60% ± 2.88% of the overall data. Importantly, the research and results presented in this paper are part of the pioneering field of the application of citizen science in the recognition of cosmic rays and, to the best of our knowledge, this analysis is performed on the largest freely available cosmic ray hit dataset.