COVID-CLNet; COVID-19 Detection with Compressive Deep Learning Approaches

One of the most serious global health threats is COVID-19 pandemic. The emphasis on increasing the diagnostic capability helps stopping its spread significantly. Therefore, to assist the radiologist or other medical professional to detect and identify the COVID-19 cases in the shortest possible time, we propose a computer-aided detection (CADe) system that uses the computed tomography (CT) scan images. This proposed boosted deep learning network (CLNet) is based on the implementation of Deep Learning (DL) networks as a complementary to the Compressive Learning (CL). We utilize our inception feature extraction technique in the measurement domain using CL to represent the data features into a new space with less dimensionality before accessing the Convolutional Neural Network. All original features have been contributed equally to the new space using a sensing matrix. Experiments performed on different compressed methods show promising results for COVID-19 detection.

Impact and Protection of GNSS Interference: Resilient Positioning in Next Generation Satellite-based Railway Train Control

Background: The Global Navigation Satellite System (GNSS) has great potentials in next generation railway train control systems. Considering the fail-safe characteristics of train control, the threat from GNSS interference may result in an increasing likelihood of outages of train positioning or even safety risks to the railway system. Objective: The interference protection solutions are investigated and demonstrated for achieving the resilient train positioning using GNSS. Methods: This paper describes the main types of GNSS interference and investigates the impact on Location Determination Unit (LDU) in the GNSS-based train control system. Specific architectures and solutions for interference detection and protection in both the position domain and measurement domain are presented. Results: Interference injection simulations are performed with both the GNSS spoofing and jamming signals, which evaluate the effects of interferences and demonstrate the protection performance of the presented solutions under GNSS attack scenarios. Conclusion: The interference protection solutions within both the position domain and measurement domain are effective and significant to mitigate the effects from the GNSS interference, which enables resilient train positioning to achieve the safe train operation.

Spatial Sampling in Monostatic Subsurface Radar Imaging

<div>This paper deals with microwave subsurface imaging obtained by a migration-like inversion scheme, for a 2D monostatic scalar configuration and a two-layered background medium. The focus is on the determination of a data sampling strategy which allows to reduce the number of required measurements and at the same time keep the same performance in the reconstructions. To this end, the measurement points are determined in order to approximate the point-spread function corresponding to the ideal continuous case (i.e., the case in which the data space is not sampled at all). Basically, thanks to suitable variable transformations the point-spread functions is recast as a Fourier-like operator and this provides insight to devise the sampling scheme. It is shown that resulting measurement spatial positions are non-uniformly arranged across the measurement domain and their number can be much lower than the one provided by some literature standard sampling criteria. The study also contains a comparison with the free space case so as to highlight the role played by the half-space that schematized the subsurface scattering scenario on the number and the locations of the measurement points. Numerical examples are also included to check the theoretical arguments.</div>

Spatial Sampling in Monostatic Subsurface Radar Imaging

<div>This paper deals with microwave subsurface imaging obtained by a migration-like inversion scheme, for a 2D monostatic scalar configuration and a two-layered background medium. The focus is on the determination of a data sampling strategy which allows to reduce the number of required measurements and at the same time keep the same performance in the reconstructions. To this end, the measurement points are determined in order to approximate the point-spread function corresponding to the ideal continuous case (i.e., the case in which the data space is not sampled at all). Basically, thanks to suitable variable transformations the point-spread functions is recast as a Fourier-like operator and this provides insight to devise the sampling scheme. It is shown that resulting measurement spatial positions are non-uniformly arranged across the measurement domain and their number can be much lower than the one provided by some literature standard sampling criteria. The study also contains a comparison with the free space case so as to highlight the role played by the half-space that schematized the subsurface scattering scenario on the number and the locations of the measurement points. Numerical examples are also included to check the theoretical arguments.</div>

Detection and Confirmation of Multiple Human Targets Using Pixel-Wise Code Aperture Measurements

Compressive video measurements can save bandwidth and data storage. However, conventional approaches to target detection require the compressive measurements to be reconstructed before any detectors are applied. This is not only time consuming but also may lose information in the reconstruction process. In this paper, we summarized the application of a recent approach to vehicle detection and classification directly in the compressive measurement domain to human targets. The raw videos were collected using a pixel-wise code exposure (PCE) camera, which condensed multiple frames into one frame. A combination of two deep learning-based algorithms (you only look once (YOLO) and residual network (ResNet)) was used for detection and confirmation. Optical and mid-wave infrared (MWIR) videos from a well-known database (SENSIAC) were used in our experiments. Extensive experiments demonstrated that the proposed framework was feasible for target detection up to 1500 m, but target confirmation needs more research.