Through-Ice Acoustic Source Tracking Using Vision Transformers with Ordinal Classification.

Ice environments pose challenges for conventional underwater acoustic localization techniques due to their multipath and non-linear nature. In this paper, we compare different deep learning networks, such as Transformers, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Vision Transformers (ViTs), for passive localization and tracking of single moving, on-ice acoustic sources using two underwater acoustic vector sensors. We incorporate ordinal classification as a localization approach and compare the results with other standard methods. We conduct experiments passively recording the acoustic signature of an anthropogenic source on the ice and analyze these data. The results demonstrate that Vision Transformers are a strong contender for tracking moving acoustic sources on ice. Additionally, we show that classification as a localization technique can outperform regression for networks more suited for classification, such as the CNN and ViTs.

Recurrent networks for direction-of-arrival identification of an acoustic source in a shallow water channel using a vector sensor.

Conventional direction-of-arrival (DOA) estimation algorithms for shallow water environments usually contain high amounts of error due to the presence of many acoustic reflective surfaces and scattering fields. Utilizing data from a single acoustic vector sensor, the magnitude and DOA of an acoustic signature can be estimated; as such, DOA algorithms are used to reduce the error in these estimations. Three experiments were conducted using a moving boat as an acoustic target in a waterway in Houghton, Michigan. The shallow and narrow waterway is a complex and non-linear environment for DOA estimation. This paper compares minimizing DOA errors using conventional and machine learning algorithms. The conventional algorithm uses frequency-masking averaging, and the machine learning algorithms incorporate two recurrent neural network architectures, one shallow and one deep network. Results show that the deep neural network models the shallow water environment better than the shallow neural network, and both networks are superior in performance to the frequency-masking average method.

Target Localization and Tracking by Fusing Doppler Differentials from Cellular Emanations with a Multi-Spectral Video Tracker.

We present an algorithm for fusing data from a constellation of RF sensors detecting cellular emanations with the output of a multi-spectral video tracker to localize and track a target with a specific cell phone. The RF sensors measure the Doppler shift caused by the moving cellular emanation and then Doppler differentials between all sensor pairs are calculated. The multi-spectral video tracker uses a Gaussian mixture model to detect foreground targets and SIFT features to track targets through the video sequence. The data is fused by associating the Doppler differential from the RF sensors with the theoretical Doppler differential computed from the multi-spectral tracker output. The absolute difference and the root-mean-square difference are computed to associate the Doppler differentials from the two sensor systems. Performance of the algorithm was evaluated using synthetically generated datasets of an urban scene with multiple moving vehicles. The presented fusion algorithm correctly associates the cellular emanation with the corresponding video target for low measurement uncertainty and in the presence of favorable motion patterns. For nearly all objects the fusion algorithm has high confidence in associating the emanation with the correct multi-spectral target from the most probable background target.

A Cluster Validity Framework Based on Induced Partition Dissimilarity.

We describe a new cluster validity framework (CVF) that compares structure in the data (in dissimilarity form) to the structure of dissimilarity matrices induced by a matrix transformation of the partition being tested. As part of this framework, we show two possible cluster validation measures: one, visual cluster validity, that that uses visual comparison and another one, correlation cluster validity, based on correlation. Unlike many existing measures, the measures we propose can be applied to crisp or soft partitions obtained by any relational or object data clustering algorithm. We illustrate the new measures and compare them to several well-known existing measures using real and artificial data sets.

An analysis of human motion detection systems use during elder exercise routines.

Human motion analysis provides motion pattern and body pose estimations. This study integrates computer-vision techniques and explores a markerless human motion analysis system. Using human-computer interaction (HCI) methods and goals, researchers use a computer interface to provide feedback about range of motion to users. A total of 35 adults aged 65 and older perform three exercises in a public gym while human motion capture methods are used. Following exercises, participants are shown processed human motion images captured during exercises on a customized interface. Standardized questionnaires are used to elicit responses from users during interactions with the interface. A matrix of HCI goals (effectiveness, efficiency, and user satisfaction) and emerging themes are used to describe interactions. Sixteen users state the interface would be useful, but not necessarily for safety purposes. Users want better image quality, when expectations are matched satisfaction increases, and unclear meaning of motion measures decreases satisfaction.

Tracking Exercise Motions of Older Adults Using Contours.

In this paper we describe the development of a novel markerless motion capture system and explore its use in documenting elder exercise routines in a health club. This system uses image contour tracking and swarm intelligence methods to track the location of the spine and shoulders during three exercises - treadmill, exercise bike, and overhead lateral pull-down. Validation results show that our method has a mean error of approximately 2 degrees when measuring the angle of the spine or shoulders relative to the horizontal. Qualitative study results demonstrate that our system is capable of providing important feedback about the posture and stability of elders while they are performing exercises. Study participants indicated that feedback from our system would add value to their exercise routines.

Measurement and data processing approach for detecting anisotropic spatial statistics of the turbulence-induced index of refraction fluctuations in the upper atmosphere.

We discuss a method of data reduction and analysis that has been developed for a novel experiment to detect anisotropic turbulence in the tropopause and to measure the spatial statistics of these flows. The experimental concept is to make measurements of temperature at 15 points on a hexagonal grid for altitudes from 12,000 to 18,000 m while suspended from a balloon performing a controlled descent. From the temperature data, we estimate the index of refraction and study the spatial statistics of the turbulence-induced index of refraction fluctuations. We present and evaluate the performance of a processing approach to estimate the parameters of an anisotropic model for the spatial power spectrum of the turbulence-induced index of refraction fluctuations. A Gaussian correlation model and a least-squares optimization routine are used to estimate the parameters of the model from the measurements. In addition, we implemented a quick-look algorithm to have a computationally nonintensive way of viewing the autocorrelation function of the index fluctuations. The autocorrelation of the index of refraction fluctuations is binned and interpolated onto a uniform grid from the sparse points that exist in our experiment. This allows the autocorrelation to be viewed with a three-dimensional plot to determine whether anisotropy exists in a specific data slab. Simulation results presented here show that, in the presence of the anticipated levels of measurement noise, the least-squares estimation technique allows turbulence parameters to be estimated with low rms error.

Experimental estimation of the spatial statistics of turbulence-induced index of refraction fluctuations in the upper atmosphere.

We present results from an experiment to estimate the parameters of homogeneous, isotropic optical turbulence in the upper atmosphere. The balloon-borne experiment made high-resolution temperature measurements at seven points on a hexagonal grid for altitudes from 12,000 to 18,000 m. From the temperature data, we obtained index of refraction fluctuations that can be used to compute a sample-based estimate for a parameterized description of the spatial autocorrelation of the turbulence. The three parameters of interest were a proportionality constant Pc, the power-law parameter alpha, and the outer scale L0. The results obtained for Pc are within the expected range and agree well with independent measurements made from a standard rising thermosonde measurement made approximately simultaneously with the data collection. Values for a were in the range 1.52 < or = alpha < or = 1.73 were observed, which are significantly less than the power law used in the Kolmogorov and von Karman models, alpha = 1.833. Values observed for L0 were in the range 5 < or = L0 < or = 19 m. Evidence that alpha may be consistently less than that used in the Kolmogorov and von Karman models likely has the most significant implications for systems that must work in or through the tropopause.