Robust Through-the-Wall Radar Image Classification Using a Target-Model Alignment Procedure
TL;DR: A technique for classifying stationary targets based on the high-range resolution profile (HRRP) extracted from 3-D TWRIs using a naive Bayesian classifier supported by principal component analysis is presented.
Abstract: A through-the-wall radar image (TWRI) bears little resemblance to the equivalent optical image, making it difficult to interpret. To maximize the intelligence that may be obtained, it is desirable to automate the classification of targets in the image to support human operators. This paper presents a technique for classifying stationary targets based on the high-range resolution profile (HRRP) extracted from 3-D TWRIs. The dependence of the image on the target location is discussed using a system point spread function (PSF) approach. It is shown that the position dependence will cause a classifier to fail, unless the image to be classified is aligned to a classifier-training location. A target image alignment technique based on deconvolution of the image with the system PSF is proposed. Comparison of the aligned target images with measured images shows the alignment process introducing normalized mean squared error (NMSE) ≤ 9%. The HRRP extracted from aligned target images are classified using a naive Bayesian classifier supported by principal component analysis. The classifier is tested using a real TWRI of canonical targets behind a concrete wall and shown to obtain correct classification rates ≥97%.
Citations
More filters
Proceedings Article•
04 May 2015TL;DR: WiTrack2.0 is presented, a multi-person localization system that operates in multipath-rich indoor environments and pinpoints users' locations based purely on the reflections of wireless signals off their bodies.
Abstract: We have recently witnessed the emergence of RF-based indoor localization systems that can track user motion without requiring the user to hold or wear any device. These systems can localize a user and track his gestures by relying solely on the reflections of wireless signals off his body, and work even if the user is behind a wall or obstruction. However, in order for these systems to become practical, they need to address two main challenges: 1) They need to be able to operate in the presence of more than one user in the environment, and 2) they must be able to localize a user without requiring him to move or change his position.
This paper presents WiTrack2.0, a multi-person localization system that operates in multipath-rich indoor environments and pinpoints users' locations based purely on the reflections of wireless signals off their bodies. WiTrack2.0 can even localize static users, and does so by sensing the minute movements due to their breathing. We built a prototype of WiTrack2.0 and evaluated it in a standard office building. Our results show that it can localize up to five people simultaneously with a median accuracy of 11.7 cm in each of the x/y dimensions. Furthermore, WiTrack2.0 provides coarse tracking of body parts, identifying the direction of a pointing hand with a median error of 12.5°, for multiple users in the environment.
307 citations
26 Apr 2014
TL;DR: This paper presents WiZ, the first multi-person centimeter-scale motion tracking system that pinpoints people’s locations based purely on RF reflections off their bodies and can track breathing motion and output the breath count of multiple people with high accuracy.
Abstract: Recently, we have witnessed the emergence of technologies that can localize a user and track her gestures based purely on radio reflections off the person’s body. These technologies work even if the user is behind a wall or obstruction. However, for these technologies to be fully practical, they need to address major challenges such as scaling to multiple people, accurately localizing them and tracking their gestures, and localizing static users as opposed to requiring the user to move to be detectable. This paper presents WiZ, the first multi-person centimeter-scale motion tracking system that pinpoints people’s locations based purely on RF reflections off their bodies. WiZ can also locate static users by sensing minute changes in their RF reflections due to breathing. Further, it can track concurrent gestures made by different individuals, even when they carry no wireless device on them. We implement a prototype of WiZ and show that it can localize up to five users each with a median accuracy of 8-18 cm and 7-11 cm in the x and y dimensions respectively. WiZ can also detect 3D pointing gestures of multiple users with a median orientation error of 8 − 16◦ for each of them. Finally, WiZ can track breathing motion and output the breath count of multiple people with high accuracy.
75 citations
TL;DR: A multipath exploitation technique using point spread functions, which associate and map back the multipath ghosts to their genuine targets, thereby increasing the effective signal-to-clutter ratio (SCR) at the genuine target locations.
Abstract: Due to several sources of multipath in through-wall radar sensing, such as walls, floors, and ceilings, there could exist multipath ghosts associated with a few genuine targets in the synthetic aperture beamformed image. The multipath ghosts are false positives and therefore confusable with genuine targets. Here, we develop a multipath exploitation technique using point spread functions, which associate and map back the multipath ghosts to their genuine targets, thereby increasing the effective signal-to-clutter ratio (SCR) at the genuine target locations. To do so, we first develop a multipath model advocating the Householder transformation, which permits modeling multiple reflections at multiple walls, and also allows for unconventional room/building geometries. Second, closed-form solutions of the multipath ghost locations assuming free space propagation are derived. Third, a nonlinear least squares optimization is formulated and initialized with these free space solutions to localize the multipath ghosts in through-wall radar sensing. The exploitation approach is general and does not require a priori assumptions on the number of targets. The free space multipath ghost locations and exploitation technique derived here may be used as is for multipath exploitation in urban canyons via synthetic aperture radar. Analytical expressions quantifying the SCR gain after multipath exploitation are derived. The analysis is validated with experimental EM results using finite-difference time-domain simulations.
74 citations
TL;DR: A signal processing strategy is proposed to track multiple extended targets in a scene by means of a wide-band monostatic through-wall radar and an extended target-tracking approach is applied to properly exploit at the tracking stage the information related to extended nature of targets.
Abstract: Tracking moving targets hidden behind visually opaque structures as building walls is a crucial issue in many surveillance, rescue, and security applications. The electromagnetic waves at the low microwave frequency range penetrate into common building materials and thereby enable the radar to expose behind the wall scene. However, due to complexity of the scattering scenario, the radar signal undergoes multipath propagation phenomena. These typically manifest themselves as environmental clutter which may impair detection and tracking of true targets. In this paper, a signal processing strategy is proposed to track multiple extended targets in a scene by means of a wide-band monostatic through-wall radar. The system collects data sets at regular time steps which are first processed by a microwave tomographic technique. Then, a detection/tracking stage is implemented in order to track the position and dynamics of targets in real time. An extended target-tracking approach is applied to properly exploit at the tracking stage the information related to extended nature of targets. The effectiveness of the proposed signal processing chain is assessed by numerical tests based on full-wave data pertaining to an indoor scenario.
70 citations
Cites background from "Robust Through-the-Wall Radar Image..."
...These include automatic target detection [19], [20], which represents a fundamental step to follow-on stages such as classification [21], [22] and tracking [23]–[25]....
[...]
TL;DR: Two multipath exploitation techniques to image a hidden target at its true location are presented under the assumptions that the locations of the reflecting walls are known and that the target multipath is resolvable and detectable.
Abstract: Multipath is exploited to image targets that are hidden due to lack of line of sight (LOS) path in urban environments. Urban radar scenes include building walls, therefore creating reflections causing multipath returns. Conventional processing via synthetic aperture beamforming algorithms do not detect or localize the target at its true position. To remove these limitations, two multipath exploitation techniques to image a hidden target at its true location are presented under the assumptions that the locations of the reflecting walls are known and that the target multipath is resolvable and detectable. The first technique directly operates on the radar returns, whereas the second operates on the traditional beamformed image. Both these techniques mitigate the false alarms arising from the multipath while simultaneously permitting the shadowed target to be detected at its true location. While these techniques are general, they are examined for two important urban radar applications: detecting shadowed targets in an urban canyon, and detecting shadowed targets around corners.
50 citations
References
More filters
Book•
01 Jan 1962
TL;DR: This chapter discusses Radar Equation, MTI and Pulse Doppler Radar, and Information from Radar Signals, as well as Radar Antenna, Radar Transmitters and Radar Receiver.
Abstract: 1 An Introduction to Radar 2 The Radar Equation 3 MTI and Pulse Doppler Radar 4 Tracking Radar 5 Detection of Signals in Noise 6 Information from Radar Signals 7 Radar Clutter 8 Propogation of Radar Waves 9 The Radar Antenna 10 Radar Transmitters 11 Radar Receiver
6,010 citations
"Robust Through-the-Wall Radar Image..." refers background or methods in this paper
...Section VII provides a summary of the work, states the conclusions drawn, and indicates future directions for this research....
[...]
...The second set of measurements was titled the classification data set, which used the dihedral, sphere, and jug of saline targets from Table IV at the positions indicated in Fig....
[...]
...The orientation of the targets in the scene was such that their angle of maximum radar cross section (RCS) was pointing directly toward the front of the scene, i.e., 0 from horizontal and forward facing....
[...]
Book•
01 Feb 1993
TL;DR: This chapter discusses how signals in Space and Time and apertures and Arrays affect Array Processing and the role that symbols play in this processing.
Abstract: 1. Introduction 2. Signals in Space and Time 3. Apertures and Arrays 4. Conventional Array Processing 5. Detection Theory 6. Estimation Theory 7. Adaptive Array Processing 8. Tracking Appendices References List of Symbols Index.
1,933 citations
"Robust Through-the-Wall Radar Image..." refers methods in this paper
...This is the type of alignment considered in conventional HRRP studies, in which the HRRP must be registered before attempting classification....
[...]
Book•
13 Apr 1999
TL;DR: Range Imaging Cross-Range Imaging SAR Radiation Pattern Generic Synthetic Aperture Radar Spotlight Synthetic aperture radar Spotlight Stripmap SyntheticAperture Radar Circular Synthetics Aperture radar Monopulse Synthetic Bperture Radar Bibliography Index.
Abstract: Range Imaging Cross-Range Imaging SAR Radiation Pattern Generic Synthetic Aperture Radar Spotlight Synthetic Aperture Radar Stripmap Synthetic Aperture Radar Circular Synthetic Aperture Radar Monopulse Synthetic Aperture Radar Bibliography Index.
1,159 citations
"Robust Through-the-Wall Radar Image..." refers background or methods in this paper
...To align the targets, this position variance of the PSF must be understood....
[...]
...The properties of the targets are detailed in Table IV, along with the abbreviated forms of their names used in the result tables....
[...]
...The second set of measurements was titled the classification data set, which used the dihedral, sphere, and jug of saline targets from Table IV at the positions indicated in Fig....
[...]
...The naive Bayesian classifier utilizes Bayes’s theorem to obtain the probability that the input data came from a given class [27]....
[...]
...The solid lines are the profiles when the target is near the imaging system, and the dashed lines, when it is far from the system....
[...]
TL;DR: In this article, the authors examine the evidence for this, both empirical, as observed in real data applications, and theoretical, summarising explanations for why this simple rule might be effective.
Abstract: Summary
Folklore has it that a very simple supervised classification rule, based on the typically false assumption that the predictor variables are independent, can be highly effective, and often more effective than sophisticated rules. We examine the evidence for this, both empirical, as observed in real data applications, and theoretical, summarising explanations for why this simple rule might be effective.
Resume
La tradition veunt qu'une regle tres simple assumant l'independance des variables predictives. une hypothese fausse dans la plupart des cas, peut etre tres efficace, souvent meme plus efficace qu'une methode plus sophistiquee en ce qui concerne l'attribution de classes a un groupe d'objets. A ce sujet, nous examinons les preuves empiriques, et les preuves theoriques, e'est-a-dire les raisons pour lesquelles cette simple regle pourrait faciliter le processus de tri.
954 citations
"Robust Through-the-Wall Radar Image..." refers background in this paper
...As discussed in Section I, the target must be detected before it can be recognized....
[...]