Vincent Radzicki

An Efficient Fusion-Based Defogging

Degradation in visibility is often introduced to images captured in poor weather conditions, such as fog or haze. To overcome this problem, conventional approaches focus mainly on the enhancement of the overall image contrast. However, because of the unspecified light-source distribution or unsuitable mathematical constraints of the cost functions, it is often difficult to achieve quality results. In this paper, a fusion-based transmission estimation method is introduced to adaptively combine two different transmission models. Specifically, the new fusion weighting scheme and the atmospheric light computed from the Gaussian-based dark channel method improve the estimation of the locations of the light sources. To reduce the flickering effect introduced during the process of frame-based dehazing, a flicker-free module is formulated to alleviate the impacts. The systematic assessments show that this approach is capable of achieving superior defogging and dehazing performance, compared with superior defogging and dehazing performance, compared with the state-of-the-art methods, both quantitatively and qualitatively.

A portable SIMO radar for through wall detection and imaging

This paper details the hardware operation, theoretical underpinnings and field validation tests of a portable, ultra-wideband, SIMO stepped frequency continuous wave sense through the wall radar system. The system is fully software defined and capable of transmitting and receiving a linearly stepped frequency waveform between 1GHz and 4GHz with reconfigurable frequency hop and sweep rates. Using such a system as the starting point, the theoretical method for extracting target range information from the receive frequency sweeps and constructing radar imagery is detailed using the pulse echo model. Combining this imaging technique along with the radar hardware yields a system capable of spatially locating targets inside of buildings from standoff distances. Experiments and verification tests of the system were conducted at the Indiana Muscatatuck Urban Training center and results are presented detailing system performance for static and moving targets inside of concrete block buildings. Radar imaging results demonstrate the systems spatial resolution and sensitivity to both moving and stationary targets.

High-resolution MIMO X-band radar for side-looking anomaly detection

Ground based, mobile surface anomaly sensing and detection is a critical area of research in explosive hazard detection as well as local situational awareness and even autonomous operations. Increasingly, achieving reliable detection is coming to rely on a suite of different (often orthogonal) sensing modalities from optical to infrared to lidar and radar. Radar is of particular interest because it offers advantages when attempting to detect obscured surface anomalies and has the potential for large observation areas. Radar’s chief disadvantage in this context is that limited physical antenna aperture degrades the spatial localization of scattering returns. This is particularly troubling in highly cluttered surface environments. To address this shortcoming of spatial localization of scattering returns, this paper discusses the use of a MIMO X-band radar system configured in a high-resolution side-looking instantiation. By configuring the system this way it can be operated in a traditional stripmap synthetic aperture mode and since it is a MIMO array it has vertical aperture allowing for three-dimensional imagery to be formed. This paper details system elements, configuration and operation of a high resolution ground based, mobile MIMO X-band radar for side-looking anomaly detection. The system operates in X-band and utilizes a digitally synthesized frequency modulated continuous waveform. The system has previously been configured for forward looking mobile anomaly detection. The work presented in this paper is concerned with synthesizer and radio frequency electronics upgrades, side-looking specific configuration issues and image formation issues. Example side-looking three-dimensional imagery is shown using canonical calibration targets.

Towards non-invasive electrocardiographic imaging using regularized neural networks

We present a new data-driven technique for non-invasive electronic imaging of cardiovascular tissues using routinely-measured body-surface electrocardiogram (ECG) signals. While traditional ECG imaging and 3D reconstruction algorithms typically rely on a combination of linear Fourier theory, geometric and parametric modeling, and invasive measurements via catheters, we show in this work that it is possible to learn the complicated inverse map, from body-surface potentials to epicardial or endocardial potentials, by exploiting the powerful approximation properties of neural networks. The key contribution here is a formulation of the inverse problem that allows historical data to be leveraged as ground-truth for training the inverse operator. We provide some initial experiments, and outline a path for extending this technique for real-time diagnostic applications.

Nonlinear electrocardiographic imaging using polynomial approximation networks

Electrocardiography is a valuable tool to aid in medical understanding and treatment of heart-related ailments, specifically atrial fibrillation (AF) and other irregular cardiac behavior. Although signs of AF will manifest in conventional electrocardiogram (ECG) recordings, interpretation and localization of AF sources require significant clinical expertise. In this vein, electrocardiographic imaging has emerged as an important medical imaging modality that provides reconstructions of the hearts electrical activity from non-invasive multi-lead body-surface ECG and anatomical x-ray computed tomography images. In this paper, we present a nonlinear inversion model for computing this mapping to improve upon the reconstruction performance of current methods. While contemporary techniques typically determine an inverse solution by discretizing and inverting an underdetermined linear system of partial differential equations governing the relationship between voltage potentials of the heart and torso, the presented technique re-casts this problem as a task in function approximation and provides a direct parameterization of the inverse operator using a polynomial neural network. That is, the outlined nonlinear inversion technique is a generalization of contemporary reconstruction techniques which allows geometrical and material parameterizations of the forward-model to be optimized using real experimental data collected from patients suffering from AF, as to better represent the inverse operator with respect to reconstruction metrics applicable to electrophysiology. The accuracy of our model is evaluated against a dataset of real-patient recordings to demonstrate its validity, and mathematical analysis is provided to support the polynomial expansion used in our inversion model.

A multistatic Doppler radar system with application to aeroecology

This paper focuses on a distributed multistatic Doppler radar system for use in aeroecology applications. The system is composed of five spatially separated receiving stations and three transmitting stations operating in offset frequency channels in S-band. The spatial and frequency diversity in the system allows for a moving targets position within the coverage area to be extracted via multilateration from its Doppler signature alone. S-band operation is chosen to allow for foliage penetration in wooded and forested areas. Simulated results show achievable position uncertainty of less than 1 m for very small cross-section targets (≤ −36dBm2) at bistatic ranges up to 300 m. System theory of operation, design considerations and performance metrics are presented using simulation as well as initial system measurements from prototype hardware.

Detection and tracking of human targets in indoor and urban environments using through-the-wall radar sensors

Radar based detection of human targets behind walls or in dense urban environments is an important technical challenge with many practical applications in security, defense, and disaster recovery. Radar reflections from a human can be orders of magnitude weaker than those from objects encountered in urban settings such as walls, cars, or possibly rubble after a disaster. Furthermore, these objects can act as secondary reflectors and produce multipath returns from a person. To mitigate these issues, processing of radar return data needs to be optimized for recognizing human motion features such as walking, running, or breathing. This paper presents a theoretical analysis on the modulation effects human motion has on the radar waveform and how high levels of multipath can distort these motion effects. From this analysis, an algorithm is designed and optimized for tracking human motion in heavily clutter environments. The tracking results will be used as the fundamental detection/classification tool to discriminate human targets from others by identifying human motion traits such as predictable walking patterns and periodicity in breathing rates. The theoretical formulations will be tested against simulation and measured data collected using a low power, portable see-through-the-wall radar system that could be practically deployed in real-world scenarios. Lastly, the performance of the algorithm is evaluated in a series of experiments where both a single person and multiple people are moving in an indoor, cluttered environment.

Standoff CW radar for through-the-wall detection of human heartbeat signatures

The design and development of a personnel radar detection system based on the identification of micro-Doppler shifts from heartbeat motion is presented here. The paper reviews the phenomenon of micro-Doppler effects measured in radar systems, and describes a specific detection scheme based on this theory. The detection performance for a general radar system is described, and specific results for a prototype system are then presented and analyzed in support of the theory. The signal processing techniques used to estimate heartbeat rates are described in detail and experimental results are provided for actual human targets demonstrating the effectiveness of the system design. The results focus on the detection capability as a function of standoff range to provide a real-world performance measure of the system.