This PDF file contains the front matter associated with SPIE Proceedings Volume 11745 including the Title Page, Copyright information, and Table of Contents.

Introduction to SPIE Defense and Commercial Sensing conference 11745: Passive and Active Millimeter-Wave Imaging XXIV

A concept for radar imaging based on multiple incoherent millimeter wave antennas is presented. The goal is to use 3D-Imaging, where the object moves along a line and passes a linear array of antennas. To reduce the hardware complexity, a novel radar principle is introduced, which does not require a coupling between TX and RX, but a reference reflector. The RX performs just a RF-power measurement for each frequency step in a SFCW setup. After preprocessing the raw data, any focusing algorithm can be used to obtain the 2D image. Multiple 2D images can be combined to 3D with backprojection.

The U.S. Army Combat Capabilities Development Command (DEVCOM) Army Research Laboratory (ARL) has been investigating the capability of using 3-D millimeter-wave synthetic aperture radar (SAR) technology to provide navigation for aircraft through terrain and obstacles in a degraded visual environment (DVE). One of the key challenges associated with this technology is the focus quality of the 3-D millimeter-wave SAR image, which requires very-high-accuracy data from the positioning measurement system for the platform motion in the 3-D space. The level of accuracy needed is currently not achievable by state-of-the-art positioning measurement systems. Thus, an effective autofocus algorithm is critical for the success of this program. In this paper, we propose an image metric-based autofocus algorithm to address the SAR image focusing challenge associated with the Ka frequency band, our MIMO radar configuration and forward-looking 3-D geometry. The algorithm models the errors in the measured radar positioning data as the rotation and translation parameters of the MIMO antenna frame in 3-D space. The algorithm solves an optimization problem that maximizes the specified SAR image metric. We evaluate the proposed algorithm and compare its performance with the Phase Gradient Autofocus (PGA) algorithm using both linear side-looking SAR and our proposed forward-looking 3-D SAR geometries.

Utilizing a generalized spotlight SAR data collection formalism described in the literature, we demonstrate a technique whereby the generalized planar flight path as well as more arbitrary collection surfaces may be analyzed in the physical scale model compact radar range environment. We briefly review the generalized planar collection geometry, its characteristics, and the implications of collecting spotlight SAR data across non-planar surfaces. Using data collected in a 305-375 GHz compact radar range, we present first results of spotlight SAR imagery modeling the generalized planar collection geometry as well as a more arbitrary geometry at X-band using 1/35th physical scale model objects. The scene observed in this work consisted of a test scene consisting of canonical scatterers as well as a scene composed of a 1/35th scale model SA-6 that exemplifies the limits of depth of focus due to non-planar motion. Our imagery results are produced from both squint and broadside collections.

Active incoherent millimeter-wave imaging is an emerging technology that combines the benefits of passive and active millimeter-wave imaging by using incoherent noise transmission and passive receive interferometric processing. In this paper, we investigate computational image processing techniques to achieve enhanced image reconstruction with a small number of antenna elements and without the need for accurate calibration. We demonstrate enhancement of images through simulation and experimental data collected with a 38-GHz active incoherent millimeter-wave imager.

Scene reconstruction through simulation with millimetre wave radar can allow for autonomous perception in poor weather, but requires an accurate forward and inverse propagation model. Ray-based light-transport is computationally efficient but generally neglects wave effects relevant to millimetre imaging. Challenges of note are the temporal coherence of signals and resulting multi-path interference glint and fade; and the spatial coherence of antennas resulting in beam patterns and phased arrays. We propose Wigner phasor signal rendering, an enhanced geometric-optics simulation framework which includes spatial and temporal coherence for modelling the response of a scene for a given input signal. This method is based on the coherent summation of waves and the representation of a wavefront as rays of constant phase. By studying the phase-space representation of a signal in separable time and space and applying appropriate transforms for propagation and interaction, we find that ray-based transport and evaluation of the transport integral yields results capturing expected wave behaviour. Using the phase-space formulation we develop three concepts for coherent signal rendering. First, the antennas radiance function is described in both position and direction via its Wigner Distribution Function. This allows for accurate and simple modelling without the imposition of beam-width or side-lobe assumptions. Second, we introduce the concept of virtual elements, locations in the phase-space of phased-arrays which contribute no energy but carry information about interference. Third, we present phasor channels and coherence kernels as a way to evaluate temporal interference as a sum of relative phases without keeping a list of all interactions. We demonstrate the validity of our framework through simulation, comparisons against the incoherent case, and measured field data.

Synthetic aperture radar (SAR) imaging has been widely used for various nondestructive evaluation (NDE) applications. The sampling strategy used to collect imaging data has great implications on the resulting image quality. The most widely-used strategies include uniform sampling and nonuniform sampling. While the former can provide relatively higher resolution and lower noise level, the latter can provide faster scanning time. However, applying uniform sampling for high resolution can be a critical issue when scanning a relatively large area. Moreover, neither of them takes target properties (e.g., depth, spatial distribution, etc.) directly into account. It has been verified that the optimum SAR resolution is target depth dependent, which means SAR intrinsically has lower resolution for targets at larger depths. This indicates that the sampling step can be accordingly increased for targets at large depths with little resolution degradation. Meanwhile, if the scene under test is relatively large and the flaws (usually just a few) are located in a relatively small region, then optimum uniform sampling over the entire large aperture, rather than a smaller area directly above the targets, may be unnecessary. Thus, first estimating target distribution density can help reduce the time in collecting imaging data. Consequently, an intelligent sampling strategy, with considerations of targets properties, is highly desired and investigated in this paper.

Lens-less millimeter-wave (mmWave) imaging of moving objects using a sparse array relies on knowledge of the relative positions between the moving object and the imaging system to enable coherent image reconstruction. However, accurate object position information is rarely available in commercial applications where the moving object, e.g. a conveyor belt or a robot, is controlled independently of the imaging system, or where the imaged objects move autonomously. This poses a significant hurdle for many commercial mmWave imaging applications. We present a video-based motion extraction approach for active mmWave imaging. The object velocity is extracted in real time from motion vectors obtained from a compressed video. This information is combined with readouts from a distance sensor to infer the position of the object at each time instant. Leveraging video-derived motion vectors enables the offloading of computational complexity of 2-D spatial correlations to highly optimized algorithms operating on camera frames. We show experimentally that the image quality of a commercial high-throughput 3-D mmWave imaging system prototype is improved significantly by this approach when the velocity of the target is unknown and time-varying. We furthermore show that image quality is also improved compared to known average motion profiles of the imaged objects. Using a lab setup with known ground truth, we show that the RMS position error is 2.5 mm over a travel length of 0.52 m. This is better than 1/8 of the wavelength at K-band (24 GHz) along the trajectory and thus sufficient to achieve excellent image quality at K-band and longer wavelengths.

Radar systems for direction of arrival (DoA) estimation have been the subject of significant research with applications ranging from security to channel sounding and automotive radars. Conventional DoA retrieval techniques rely on an array based system architecture as the receiving unit, typically synthesized at the Nyquist limit. This classical array based approach makes it necessary to collect the received radar signals from multiple channels, and process it using DoA estimation algorithms to retrieve the DoA information of incoming far-field sources. A challenge with this multi-pixel approach is that, as the operating frequency is increased, the number of antennas (and hence the number of data acquisition channels) also increases. This can result in a rather complex system architecture at the receiver unit, especially at millimetre-wave and submillimetre-wave frequencies. As an enabling technology for the compressing sensing paradigm, a single-pixel based coded aperture can substantially simplify the physical hardware layer for DoA estimation. A significant advantage of this technique is that the received data from the source is compressed into a single channel, circumventing the necessity to have array-based multiple channels to retrieve the DoA information. In this work, we present a passive compressive sensing radar technique for DoA estimation using a single-frequency, dynamically reconfigurable wave-chaotic metasurface antenna as a receiver. We demonstrate that using spatiotemporarily incoherent measurement modes generated by the coded programmable metasurface aperture to encode and compress source generated far-field incident waves into a single channel, we can retrieve high fidelity DoA patterns from compressed measurements.

Computational millimetre-wave (mmW) imaging and machine learning have followed parallel tracks since their inception. Recent developments in computational imaging (CI) have significantly improved the imaging capabilities of mmW imaging systems. Machine learning algorithms have also gained huge popularity among researchers in the recent past with several approaches being investigated to make use of them in imaging systems. One such algorithm, image classifier, has gained significant traction in applications such as security screening and traffic surveillance. In this article, we present the first steps towards a machine learning integrated CI physical model for image classification at mmW frequencies. The dataset used for training CI system is generated using the developed single-pixel CI forward-model, eliminating the need for traditional raster-scanning based imaging techniques.

Reducing the latency of electromagnetic imaging is a crucial objective for applications such as security screening, autonomous driving, and touchless human-machine interaction. In that respect, a fundamental caveat with conventional compressed sensing techniques is that initially all information is indiscriminately multiplexed across a diverse set of measurement modes, and only during data processing one begins to select the information that is relevant to the task (e.g. concealed weapon detection). In order to only acquire relevant information in the first place, and hence drastically reduce the number of necessary measurements, the “learned sensing” paradigm suggests to interpret reconfigurable measurement hardware (e.g. a dynamic metasurface aperture) as a trainable physical layer. The latter can be directly integrated into the machine-learning pipeline used on the data processing side such that one can jointly optimize physical weights (measurement settings) and digital weights (processing network). We discuss our recent numerical and experimental investigations of this new approach to electromagnetic imaging. Our results show that a considerable reduction of the number of scene illuminations is possible by using learned illumination patterns as opposed to conventional patterns (random, orthogonal or principal scene components). At the same time, we find that there are no intuitive explanations for the learned patterns. We also clarify whether the resolution of sub-wavelength features in the scene is limited by the conventional diffraction limit.

A CNT has a very low light density, about half that of aluminum, but its strength is about 20 times that of steel. Its current density resistance is more than 1000 times that of copper, and it also has higher thermal conductivity than copper. Although it is an excellent material that can be expected to be used in various fields, using it effectively proves to be a difficult task. We have succeeded in developing a flexible and stretchable CNT paint with the aim of applying this excellent material to DEA electrodes. This CNT paint can be easily applied to various materials by using a spray or other methods and since it can expand and contract, it can be applied to easily deformable materials such as polymers, wood, paper, and resin. It is possible to add the characteristics of CNTs to various materials. One of the characteristics of CNTs that has been attracting attention in recent years is the radio wave absorption effect. As the understanding of this process is expected to grow ever more important, the purpose of this paper is to verify the effect of radio wave absorption, one of the methods of utilizing CNTs, in the range from single to high double digit GHz bands.

Non-paper inclusions inside paper and plastic document mailing envelopes present both economic loss and security concerns for shipping carriers. Such contraband often goes unnoticed unless an envelope is physically opened by a human, which is infeasible given a global shipping volume in the tens of millions of envelopes per day. Millimeter waves (mmWaves) penetrate most non-metallic packaging materials, enabling the detection of anomalous nonpaper items within a stack of documents. At the same time, the non-ionizing nature of mmWave energy enables the safe use of mmWave imaging in close proximity to human workers without a requirement for shield barriers. We demonstrate a high-throughput K-Band (24 GHz) mmWave imaging system used to scan envelopes and thin packages transiting a conveyor belt. This imaging system is capable of supporting conveyor speeds of up to 3 m/s and enables non-destructive imaging inside sealed envelopes. We also present an automated screening algorithm that uses a logistic regression approach to detect anomalies among the expected paper documents. Automatic anomaly detection removes the human from the equation and allows for high-throughput diversion of suspect envelopes for secondary screening. In this work, we investigate mmWave detection of non-paper inclusions such as metalized plastic and metal items among paper documents in paper, cardboard, and Tyvek envelopes, as well as padded bubble packs. We achieve resolution better than 1 cm in the plane of the envelope, allowing for identification of sub-cm3 anomalies, and demonstrate automated first-pass flagging of suspect envelopes.

The COVID-19 pandemic has created massive socio-economic disruptions globally. As it is certain that this pandemic is not the last one that the World is to experience, there are ongoing massive efforts to strengthen the resilience of today's globalised economy against the effects of future pandemics. One such measure, which has deployed widely, is the use of infrared cameras to detect febrile persons at e.g. border crossings. However, infrared thermography suffers from a number of shortcomings, which lead to high sensitivity but very poor specificity of detection. In this paper we outline our proposed method of full-body sub-mm wave thermography. The method builds on the fact that most clothing materials are relatively transmissive to submm-waves, while the skin emissivity is high, giving access to full-body thermal images. Highly specific detection of fever is possible thanks to the human body's physiological response to infections. Initial results from a small number of febrile persons will be presented.

For numerous years, INO has been developing active video rate THz imaging systems operating in the 250-750 GHz band. These systems are designed for use in application fields such as security and industrial inspection. Although such systems are already deployed in the field, standard procedures for determining key metrics of their performances such as resolution and SNR are still work in progress. To support and validate the ongoing development of our systems, proper characterization methods are needed. This article describes our development on the use of various resolution targets and measurement procedures for characterizing our FPA-based THz active imaging system prototypes operated in reflection mode (collecting energy reflected by the observed scene). We analyze and discuss the results obtained with different resolution targets such as bar charts, Siemens chart, slanted edge and point sources. The repeatability and applicability of the methods are assessed by repeating the measurement procedure and analyzing the measurement discrepancies. Our results are compared to theoretical expectations when available.

This work describes the characterization of a powder threat material and the development of a dielectric simulant for a millimeter wave imaging system. A resonant cavity measurement system was designed and implemented to characterize the permittivity of the powder threat material and the candidate simulant formulation for the frequency range of the millimeter wave imaging system. The Resonant Cavity Measurement System determines the complex permittivity of the material as a function of density. A reflection coefficient is calculated at each frequency from a Fresnel reflection and transmission model developed in the electromagnetic signatures of explosives laboratory. The model allows different backing materials (skin and air) and material thicknesses to be examined. Finally, the Houston Criterion is applied to determine whether the matched simulant material and its paired explosive material can be distinguished from each other by the millimeter wavelength imaging system. The explosive material and the simulant material were found to be a dielectric match for use with the millimeter wavelength imaging system of interest with both backing materials (skin and air). The Houston criterion for resolvability was not exceeded for any thickness for either backing material. Therefore, the two materials are indistinguishable by the millimeter wavelength imaging system of interest.

Over the last decade, significant progress has been made in the development of Terahertz (THz) imagers to satisfy the growing interest for see-through devices for different market applications. The noise-equivalent power (NEP) is a widely accepted figure of merit used to compare the sensitivity performance of detectors. However, with no widely recognized standard for NEP, it is often difficult to have a fair comparison between different sensors. Having a clear understanding of the characterization method used to calculate this important metric will lead to better estimation of the performances that could be expected from an imaging device. There is some confusion regarding whether NEP should be expressed in terms of power (W) or power by spectral density (W/Hz^1/2). The difference between the two expressions is the normalization of the first by the square root of the detector’s equivalent noise bandwidth (ENBW). By properly defining the ENBW for a specific sensor, the translation between the two is then consistent. This paper presents the NEP characterization of INO’s Microxcam-384i camera over a wide frequency range. A description of the measurement setup is provided, as well as the details of the analysis method, including the estimation of the ENBW. Finally, values for the NEP using both expressions are provided for wavelengths between 70μm (4.5 THz) and 1.5mm (198 GHz), demonstrating the broadband sensitivity of the camera.

Active millimeter-wave imaging is in widespread use for security screening and other applications. The Pacific Northwest National Laboratory (PNNL) has developed a variety of microwave and millimeter-wave imaging systems and technology, including the cylindrical imaging technology that forms the basis of the L3/Leidos ProVision system. Since 2016, PNNL has been actively participating in a working group that is developing a proposed American National Standards Institute (ANSI) standard (N42.59) that will be used to evaluate and verify performance of active millimeterwave imaging systems used for security screening of humans. The standard is developing image quality tools (IQTs) that will be used to assess a variety of imaging metrics, such as lateral resolution, contrast, and depth resolution. Depth resolution is vital for high-performance microwave and millimeter-wave imaging because it enables precise focusing over a full 3D volume, and allows for differentiation of reflections from multiple surfaces, such as a layer of clothing over the human body. In this paper, depth resolution is analyzed using theoretical simulations and experimental 3D imaging studies. Presented results examine depth resolution using IQTs developed from a thin partially transparent film placed in front of a metallic surface, creating reflections that are laterally aligned and at variable separation. Coherence between these reflections is investigated as it complicates the interpretation of the imaging results.

The JHU/APL-developed Virtual Advanced Imaging Technology (VAIT) models the electromagnetic reflections associated with current and future airport checkpoint scanners, producing images suited for studying threat detection phenomenology and detection algorithm training/testing. It employs an efficient GPU-based electromagnetic simulation engine that computes the field reflected by computer-modeled humans, clothing, and concealed threat items with proper material properties. Optimized image-reconstruction algorithms then use reflectivity data to render synthetic AIT images of similar quality as those produced by vendor AIT systems. The VAIT has been designed to emulate various AIT architectures, including traditional pose AIT systems and walkthrough systems.

Active millimeter and microwave imaging techniques can be used to create a high-resolution 3D image volume of a target’s reflectivity. Millimeter-wave imaging is commonly used for personnel security screening and numerous other applications. Backprojection based image reconstruction techniques form a 3D complex-valued volume. The complex-valued volume is commonly reduced to a real-valued volume by taking the magnitude. For anomaly detection and cross image registration of an object it is beneficial to generate an accurate representation of an object’s surface. Extracting a smooth and accurate surface from a magnitude only image is challenging. One difficulty is the magnitude image has limited resolution in the depth direction which normally limits precision to a moderate fraction of the depth resolution. Another difficulty is that the magnitude of the image depends heavily on the geometry and orientation of the object being imaged. The phase information in a complex-valued image volume provides a means to decouple the magnitude of the image from the geometry of an object and provide precision much finer than the depth resolution would indicate. This enables the generation of a smooth and accurate point cloud representation of the surface of an imaged object. A method to extract a point cloud from the phase information in a 3D complex-valued millimeter-wave image volume is developed and results with simulated and experimental data are presented