This PDF file contains the front matter associated with SPIE Proceedings Volume 8734, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

The human’s Micro-Doppler signatures resulting from breathing, arm, foot and other periodic motion can provide valuable information about the structure of the moving parts and may be used for identification and classification purposes. In this paper, we carry out simulate with FDTD method and through wall experiment with UWB radar for human being’s periodic motion detection. In addition, Advancements signal processing methods are presented to classify and to extract the human’s periodic motion characteristic information, such as Micro-Doppler shift and motion frequency. Firstly, we apply the Principal Component Analysis (PCA) with singular value decomposition (SVD) to denoise and extract the human motion signal. Then, we present the results base on the Hilbert-Huang transform (HHT) and the S transform to classify and to identify the human’s micro-Doppler shift characteristics. The results demonstrate that the combination of UWB radar and various processing methods has potential to detect human’s Doppler signatures effectively.

Wideband radar provides a significant improvement over traditional narrowband radars for micro-Doppler analysis because the high bandwidth can be used to separate many of the signals in range, allowing a simpler decomposition of the micro-Doppler signals. Recent wideband radar work has focused on micro-Doppler, but there is a point where the narrowband approach used to analyze the micro-Doppler signals breaks down. The effect is shown to be independent of frequency, but the error relative to the bandwidth is shown to be inversely proportional to the frequency and proportional to the velocity of the subject. This error can create a smearing effect in the observed Doppler if it is not corrected, leading to reduced signal-to-noise and the appearance of more diffuse targets in Doppler space. In range-space, wideband data can also break the subject into several range bins, affecting the observed signal to noise ratio. The possible applications of wideband micro-Doppler radar are also shown, including the separation of arm movement from human motion which implies that the arms are not encumbered.

In this paper, we present a novel, standalone ultra wideband (UWB) micro-Doppler radar sensor that goes beyond simple range or micro-Doppler detection to combined range-time-Doppler frequency analysis. Moreover, it can monitor more than one human object in both line-of-sight (LOS) and through wall scenarios, thus have full human objects tracking capabilities. The unique radar design is based on narrow pulse transceiver, high speed data acquisition module, and wideband antenna array. For advanced radar post-data processing, joint range-time-frequency representation has been performed. Characteristics of human walking activity have been analyzed using the radar sensor by precisely tracking the radar object and acquiring range-time-Doppler information simultaneously. The UWB micro-Doppler radar prototype is capable of detecting Doppler frequency range from -180 Hz to +180 Hz, which allows a maximum target velocity of 9 m/s. The developed radar sensor can also be extended for many other applications, such as respiration and heartbeat detection of trapped survivors under building debris.

In this paper, a novel and comprehensive measurement approach is proposed for the detection and analysis of human motion signature. The approach combines theoretical concepts and tools of micro-Doppler theory, image processing, and human modeling, in a original way. The attention is primarily focused on the description of the most meaningful parameters influencing the accuracy of the obtained signature. The ultimate purpose is to provide a framework through which organizing, comparing, and merging future research activities, ideas and results in the field of human motion signature analysis for security, health and disaster recovery purposes. Some simulation and experimental results underlying the feasibility and effectiveness of the measurement approach are also summarized and analyzed.

With the advances in radar technology, there is an increasing interest in automatic radar-based human gait identification. This is because radar signals can penetrate through most dielectric materials. In this paper, an image-based approach is proposed for classifying human micro-Doppler radar signatures. The time-varying radar signal is first converted into a time-frequency representation, which is then cast as a two-dimensional image. A descriptor is developed to extract micro-Doppler features from local time-frequency patches centered along the torso Doppler frequency. Experimental results based on real data collected from a 24-GHz Doppler radar showed that the proposed approach achieves promising classification performance.

The identification and classification of human motions has become a popular area of research due to its broad range of applications. Knowledge of a person's movements can be a useful tool in surveillance, security, military combat, search and rescue operations, and the medical fields. Classification of common stationary human movements has been performed under various scenarios for two different micro-Doppler radar systems: S-band radar and millimeter-wave (mm-wave) radar. Each radar system has been designed for a specific scenario. The S-band radar is intended for through-the-wall situations at close distances, whereas the mm-wave radar is designed for long distance applications and also for through light foliage. Here, the performance of these radars for different training scenarios is investigated. The S-band radar will be analyzed for classification without a wall barrier, through a brick wall, and also through a cinder block wall. The effect of a wall barrier on micro-Doppler signatures will be briefly discussed. The mm-wave radar will be analyzed for classification at distances of 30, 60, and 91 meters.

The human micro-Doppler signature is a unique signature caused by the time-varying motion of each point on the human body, which can be used to discriminate humans from other targets exhibiting micro-Doppler, such as vehicles, tanks, helicopters, and even other animals. Classification of targets based on micro-Doppler generally involves joint timefrequency analysis of the radar return coupled with extraction of features that may be used to identify the target. Although many techniques have been investigated, including artificial neural networks and support vector machines, almost all suffer a drastic drop in classification performance as the aspect angle of human motion relative to the radar increases. This paper focuses on the use of radar networks to obtain multi-aspect angle data and thereby ameliorate the dependence of classification performance on aspect angle. Knowledge of human walking kinematics is exploited to generate a fuse spectrogram that incorporates estimates of model parameters obtained from each radar in the network. It is shown that the fused spectrogram better approximates the truly underlying motion of the target observed as compared with spectrograms generated from individual nodes.

The U.S. Army Research Laboratory (ARL) supports the development of classification, detection, tracking, and localization algorithms using multiple sensing modalities including acoustic, seismic, E-field, magnetic field, PIR, and visual and IR imaging. Multimodal sensors collect large amounts of data in support of algorithm development. The resulting large amount of data, and their associated high-performance computing needs, increases and challenges existing computing infrastructures. Purchasing computer power as a commodity using a Cloud service offers low-cost, pay-as-you-go pricing models, scalability, and elasticity that may provide solutions to develop and optimize algorithms without having to procure additional hardware and resources. This paper provides a detailed look at using a commercial cloud service provider, such as Amazon Web Services (AWS), to develop and deploy simple signal and image processing algorithms in a cloud and run the algorithms on a large set of data archived in the ARL Multimodal Signatures Database (MMSDB). Analytical results will provide performance comparisons with existing infrastructure. A discussion on using cloud computing with government data will discuss best security practices that exist within cloud services, such as AWS.

One of the major scientific thrusts from recent years has been to try to harness quantum phenomena to dramat­ ically increase the performance of a wide variety of classical devices. These advances in quantum information science have had a considerable impact on the development of photonic-based quantum sensors. Even though quantum radar and quantum lidar remain theoretical proposals, preliminary results suggest that these sensors have the potential of becoming disruptive technologies able to revolutionize reconnaissance systems. In this paper we will discuss how quantum entanglement can be exploited to increase the radar and lidar signature of rectangular targets. In particular, we will show how the effective visibility of the target is increased if observed with an entangled multi-photon quantum sensor.

Field and laboratory measurements of Light Detection and Ranging (LIDAR) sensor interactions with dust have been performed for two types of common ground-based LIDAR sensors. A strong correlation (R2 > 0.95) between the probability for a return from the dust and the optical depth of the dust was found in the analysis. Based on the experimental correlation, a probabilistic model for LIDAR interactions with dust was developed and verified in field experiments. Finally, the model was integrated into a high-fidelity ray-tracing simulation of LIDAR systems.

Deep-ultraviolet resonance Raman spectroscopy (DUVRRS) is a promising approach to stand-off detection of explosive traces due to large Raman cross-section and background free signatures. In order to design an effective sensor, one must be able to estimate the signal level of the DUVRRS signature for solid-phase explosive residues. The conventional approach to signal estimation uses scattering cross-sections and molar absorptivity, measured on solutions of explosives dissolved in an optically-transparent solvent. Only recently have researchers started to measure solid-state cross-sections. For most solid-phase explosives and explosive mixtures, neither the DUV Raman scattering cross sections nor the optical absorption coefficient are known, and they are very difficult to separately measure. Therefore, for a typical solid explosive mixture, it is difficult to accurately estimate Raman signal strength using conventional approaches. To address this issue, we have developed a technique to measure the Raman scattering strength of optically-thick (opaque) materials, or “Raman Albedo”, defined as the total power of Raman-scattered light per unit frequency per unit solid angle divided by the incident power of the excitation source. We have measured Raman Albedo signatures for a wide range of solid explosives at four different DUV excitation wavelengths. These results will be presented, and we will describe the use of Raman Albedo measurements in the design and current construction of a novel stand-off explosive sensor, based on dual-excitation-wavelength DUVRRS.

SiOnyx has developed infrared enhanced CMOS image sensors leveraging a proprietary ultrafast laser semiconductor process technology. This technology demonstrates 10 fold improvements in infrared sensitivity over incumbent imaging technology while maintaining complete compatibility with standard CMOS image sensor process flows. Furthermore, these sensitivity enhancements are achieved on a focal plane with state of the art noise performance of 2 electrons/pixel. The focal plane is color enabled but high transmission of near infrared light allows for near infrared imaging from 850 to 1200 as well. The quantum efficiency enhancements have significant performance benefits in imaging 1064nm laser light as well as 850nm imaging of iris signatures for improved biometric identification.

To evaluate the performance of speaker recognition systems, a detection cost function defined as a weighted sum of the probabilities of type I and type II errors is employed. The speaker datasets may have data dependency due to multiple uses of the same subjects. Using the standard errors of the detection cost function computed by means of the two-layer nonparametric two-sample bootstrap method, a significance test is performed to determine whether the difference between the measured performance levels of two speaker recognition algorithms is statistically significant. While conducting the significance test, the correlation coefficient between two systems’ detection cost functions is taken into account. Examples are provided.

The capability to detect, observe, and positively identify people at a distance is important to numerous security and defense applications. Traditional solutions for human detection and observation include long-range visible imagers for daytime and thermal infrared imagers for night-time use. Positive identification, through computer face recognition, requires facial imagery that can be repeatably matched to a database of visible facial signatures (i.e. mug shots). Nighttime identification at large distance is not possible with visible imagers, due to lack of light, or with thermal infrared imagers, due to poor correlation with visible facial imagery. An active-SWIR imaging system was developed that is both eye-safe and invisible, capable of producing close-up facial imagery at distances of several hundred meters, even in total darkness. The SWIR facial signatures correlate well to visible signatures, allowing for biometric face recognition night or day. Night-time face recognition results for several distances will be presented. Human detection and observation results at larger distances will also be presented. Example signatures will be presented and discussed.

Human motion analysis is a task of increasing importance in several modern application fields, such as in medicine, avionics, security, and disaster recovery. In this paper, the use of Hough transform is considered and discussed in the scenario of human motion analysis. In particular, the influence of some transform parameters is investigated with the aim of improving Hough transform set-up when used in a measurement approach for human motion analysis. To this purpose, Hough transform has been applied to a set of results obtained by exploiting a suitable measurement system developed by the same authors, in the specific case of ultrasound waves. Such results have been obtained by using some reference images in the form of spectrograms achieved by using the system along with a purposely developed reference target emulating some human body movements. The results show that the considered measurement system, and more generally the human motion analysis and detection system, can be optimized by a proper set-up and use of Hough transform algorithm.

This research highlights the results obtained from applying the method of inverse kinematics, using Groebner basis theory, to the human gait cycle to extract and identify lower extremity gait signatures. The increased threat from suicide bombers and the force protection issues of today have motivated a team at Air Force Institute of Technology (AFIT) to research pattern recognition in the human gait cycle. The purpose of this research is to identify gait signatures of human subjects and distinguish between subjects carrying a load to those subjects without a load. These signatures were investigated via a model of the lower extremities based on motion capture observations, in particular, foot placement and the joint angles for subjects affected by carrying extra load on the body. The human gait cycle was captured and analyzed using a developed toolkit consisting of an inverse kinematic motion model of the lower extremity and a graphical user interface. Hip, knee, and ankle angles were analyzed to identify gait angle variance and range of motion. Female subjects exhibited the most knee angle variance and produced a proportional correlation between knee flexion and load carriage.

While published literature of the optical properties of human skin is prevalent for the visible region, data are sparse in the ultraviolet and shortwave infrared. Spectral imaging has expanded from primarily an earth remote sensing tool to a range of applications including medicine and security applications, as examples. These emerging applications will likely benefit from exemplar data of human skin spectral signatures that can be used in designing and testing spectral imaging systems. This paper details an initial study of the reflectance properties over the spectral range of the ultraviolet to the shortwave infrared. A commercial spectrophotometer was used to collect the directional-hemispherical reflectance of each participant’s skin from 250 nm to 2500 nm. The measurements are directly traceable to the national scales of reflectance and include estimated measurement uncertainties. The portion of skin under test was located on the participant’s forearm and was approximately 5 mm in diameter. The results provided in this study serve as one point of reference for the optical properties of skin that in turn will aid in the development of physical and digital tissue phantoms.

The ability to accurately detect a target of interest in a hyperspectral imagery (HSI) is largely dependent on the spatial and spectral resolution. While hyperspectral imaging provides high spectral resolution, the spatial resolution is mostly dependent on the optics and distance from the target. Many times the target of interest does not occupy a full pixel and thus is concealed within a pixel, i.e. the target signature is mixed with other constituent material signatures within the field of view of that pixel. Extraction of spectral signatures of constituent materials from a mixed pixel can assist in the detection of the target of interest. Hyperspectral unmixing is a process to identify the constituent materials and estimate the corresponding abundances from the mixture. In this paper, a framework based on non-negative matrix factorization (NMF) is presented, which is utilized to extract the spectral signature and fractional abundance of human skin in a scene. The NMF technique is employed in a supervised manner such that the spectral bases of each constituent are computed first, and then these bases are applied to the mixed pixel. Experiments using synthetic and real data demonstrate that the proposed algorithm provides an effective supervised technique for hyperspectral unmixing of skin signatures.

The SDA (Spectral Dynamics Analysis) method is used for the detection and identification of the PWM C4 explosive with the surface having inhomogeneity, caused by action of the sandpaper with different grit on the explosive surface, or with the surface having various curvature of its surface. We show that the SDA-method is good tool for the detection and identification of the explosive using THz signal reflected from the PWM C4 explosive. We propose (see as well [24]) integral criteria for the identification of substances. These criteria allow to detect the explosive despite an influence of its shape on the THz spectrum. Proposed assessments and algorithms for computation of the identification probability show both high probability of the substance identification and a reliability of realization in practice.

A standoff multivariate calibration for detection of highly energetic materials (HEM) using Fourier transform infrared spectroscopy is presented in this report. The procedure consists in standoff sensing at 1 m distance and the variation of three parameters of detection. The first variable considered was the angular dependence: 0° to 45‡ from source-target with respect to alignment of target-detector. The second variable consisted on the use of several surfaces on which the material was deposited. The substrates used were polished aluminum and anodized aluminum. The third variable studied was the dependence on some specific analyte loading surface concentration: from 10 μg/cm² to200 μg/cm². The HEM used in this work was PETN, synthesized in our lab. Calibration curves were based on the use of chemometrics routines such as partial least squares (PLS) regression analysis. This algorithm was used to evaluate the impact of the angular dependence about the limits of detection of different HME loadings on aluminum substrates.