Vibrometry offers the potential to classify a target based on its vibration spectrum. Signal processing is necessary for extracting features from the sensing signal for classification. This paper investigates the effects of fundamental frequency normalization on the end-to-end classification process [1]. Using the fundamental frequency, assumed to be the engine’s firing frequency, has previously been used successfully to classify vehicles [2, 3]. The fundamental frequency attempts to remove the vibration variations due to the engine’s revolution per minute (rpm) changes. Vibration signatures with and without fundamental frequency are converted to ten features that are classified and compared. To evaluate the classification performance confusion matrices are constructed and analyzed. A statistical analysis of the features is also performed to determine how the fundamental frequency normalization affects the features. These methods were studied on three datasets including three military vehicles and six civilian vehicles. Accelerometer data from each of these data collections is tested with and without normalization.
In vehicle target classification, contact sensors have frequently been used to collect data to simulate laser vibrometry data. Accelerometer data has been used in numerous literature to test and train classifiers instead of laser vibrometry data [1] [2]. Understanding the key similarities and differences between accelerometer and laser vibrometry data is essential to keep progressing aided vehicle recognition systems. This paper investigates the contrast of accelerometer and laser vibrometer data on classification performance. Research was performed using the end-to-end process previously published by the authors to understand the effects of different types of data on the classification results. The end-to-end process includes preprocessing the data, extracting features from various signal processing literature, using feature selection to determine the most relevant features used in the process, and finally classifying and identifying the vehicles. Three data sets were analyzed, including one collection on military vehicles and two recent collections on civilian vehicles. Experiments demonstrated include: (1) training the classifiers using accelerometer data and testing on laser vibrometer data, (2) combining the data and classifying the vehicle, and (3) different repetitions of these tests with different vehicle states such as idle or revving and varying stationary revolutions per minute (rpm).
This paper evaluates and expands upon the existing end-to-end process used for vibrometry target classification and identification. A fundamental challenge in vehicle classification using vibrometry signature data is the determination of robust signal features. The methodology used in this paper involves comparing the performance of features taken from automatic speech recognition, seismology, and structural analysis work. These features provide a means to reduce the dimensionality of the data for the possibility of improved separability. The performances of different groups of features are compared to determine the best feature set for vehicle classification. Standard performance metrics are implemented to provide a method of evaluation. The contribution of this paper is to (1) thoroughly explain the time domain and frequency domain features that have been recently applied to the vehicle classification using laser-vibrometry data domain, (2) build an end-to-end classification pipeline for Aided Target Recognition (ATR) with common and easily accessible tools, and (3) apply feature selection methods to the end-to-end pipeline. The end-to-end process used here provides a structured path for accomplishing vibrometry-based target identification. This paper will compare with two studies in the public domain. The techniques utilized in this paper were utilized to analyze a small in-house database of several different vehicles.
Holographic Aperture Ladar (HAL) is an intriguing variant of Synthetic Aperture Ladar (SAL). As with
conventional SAL, HAL systems seek to increase cross-range scene resolution by synthesizing a large effective
aperture through the motion of a smaller receiver, and through the subsequent proper phasing and correlation of the
detected signals in post-processing. Unlike in conventional SAL, however, holographic aperture ladar makes use of
a two-dimensional translating sensor array, not simply a translating point detector. In real world applications less
than ideal conditions will be detrimental to final image quality. As the HAL transform requires precise knowledge
of each data collection site in order to properly phase a possibly large collection of coherent sub-images, laser pulse
jitter and system platform vibration are two factors that may result in non-optimum final image quality. To examine
these effects, we first define the following metrics which, in part, quantify final image quality: cross-range
resolution (ΔCR); peak-to-integrated-side-lobe-ratio (PISLR); peak-to-side-lobe-ratio (PSLR); and, pupil plane
RMS wavefront error. We then numerically examine the effects of data collection site uncertainty in a HAL system
via Monte Carlo simulation. In our model we consider only a single point object, though we use otherwise realistic
parameters for sub-aperture diameter, range, wavelength, etc. The effects of positional uncertainty on the image
quality metrics are then calculated, and the results compared to ideal expectations. We will present characteristic
results for several different synthetic aperture diameters and will identify regions of diffraction-limited performance
by considering Marechal's well known λ/14 RMS wavefront error criterion.
We show through numerical modeling that the range resolution of a multi-band, sparse frequency CW-LFM chirped
signal has an effective bandwidth related to the modulation bandwidth and the band frequency offsets of all bands. The
range resolution predicted from the effective bandwidth of our sparse CW-LFM signal is comparable to that of standard
continuous bandwidth CW-LFM signals. We also discuss unique issues that arise from the use of sparse frequency CWLFM
chirped signals, such as ambiguity and peak to side-lobe ratio fluctuations, and how they are related to the multiple
frequency components of the signal.
KEYWORDS: Modulation transfer functions, Spatial frequencies, Imaging systems, Image resolution, Numerical simulations, Optical transfer functions, Optical imaging, Signal attenuation, Point spread functions, Signal to noise ratio
Sparse aperture imaging systems are capable of producing high resolution images while maintaining an overall light
collection area that is small with respect to a fully filled aperture yielding the same resolution. However, conventional
sparse aperture systems pay the penalty of reduced contrast at
mid-band spatial frequencies.
The modulation transfer function (MTF), or normalized autocorrelation, provides a quantative measure of both the
resolution and contrast of an optical imaging system. Numerical MTF calculations were thus used to examine mid-band
contrast recovery through the systematic increase of autocorrelation redundancy in a Golay-9 sparse array.
In a Golay-9 sparse aperture arrangement, three sets of three
sub-apertures can be shown to lie at unique radii from the
center of the array. In order to increase the mid-frequency contrast we then have two options. The first, and most
influential, is to increase the size of the sub-apertures located at the intermediate radius from the array origin. This
directly increases autocorrelation redundancy at mid-band frequencies. The second option, though less effective, is to
increase the relative mid-band frequency response by attenuating the outer most sub-apertures.
We will demonstrate that by increasing the diameters of the mid-radii sub-apertures, mid-band contrast can be increased
by over 45%, compared to uniform sub-aperture diameter arrays. We will also demonstrate that attenuating the outer
most sub-apertures can further increase mid-band contrast recovery, but only by less than 1%. The effects on array fill
factor will also be discussed.
New results of the (temperature) refractive index structure parameter (CT2), Cn2 are presented from fast response sensor observations near the ground and also using a kite/tethered blimp platform and an aircraft, at the Edward Air Force Base in Mojave Desert, California. Additional optical measurements include near-ground scintillation observations over horizontal paths. Atmospheric turbidity were also calculated from direct beam solar radiation measurements using pyrheliometer. Comparisons were made of the observed profiles of refractive index structure parameters (Cn2) with theoretical modeled profiles, and two derived quantities such as transverse coherence length (r0) and isoplanatic angle (θ0) for a slant path are discussed. All of these parameters are the major indicators of turbulence and are important to design an aircraft or space-craft-based free-space laser communication and high resolution optical synthetic-aperture imaging systems. Non-isotropic turbulence observations from some of the data will be pointed out. Probability density functions (PDF) of the distribution of Cn2 will be described using histograms. Fundamental limits imposed by atmospheric effects in high data rate communication and optical synthetic-aperture imaging systems will be discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.