No description available.

Processing hyperspectral image data can be computationally expensive and difficult to employ for real-time applications due to its extensive spatial and spectral information. Further, applications in which computational resources may be limited can be hindered by the volume of data that is common with airborne hyperspectral image data. This paper proposes utilizing band selection to down-select the number of spectral bands to consider for a given classification task such that classification can be done at the edge. Specifically, we consider the following state of the art band selection techniques: Fast Volume-Gradient-based Band Selection (VGBS), Improved Sparse Subspace Clustering (ISSC), Maximum-Variance Principal Component Analysis (MVPCA), and Normalized Cut Optimal Clustering MVPCA (NC-OC-MVPCA), to investigate their feasibility at identifying discriminative bands such that classification performance is not drastically hindered. This would greatly benefit applications where time-sensitive solutions are needed to ensure optimal outcomes. In this research, an NVIDIA AGX Xavier module is used as the edge device to run trained models on as a simulated deployed unmanned aerial system. Performance of the proposed approach is measured in terms of classification accuracy and run time.

The human brain is unquestionably the most complex organ of the body as it controls and processes its movement and senses. A healthy brain is able to generate responses to the signals it receives, and transmit messages to the body. Some neural disorders can impair the communication between the brain and the body preventing the transmission of these messages. Brain Computer Interfaces (BCIs) are devices that hold immense potential to assist patients with such disorders by analyzing brain signals, translating and classifying various brain responses, and relaying them to external devices and potentially back to the body. Classifying motor imagery brain signals where the signals are obtained based on imagined movement of the limbs is a major, yet very challenging, step in developing Brain Computer Interfaces (BCIs). Of primary importance is to use less data and computationally efficient algorithms to support real-time BCI. To this end, in this thesis we explore and develop algorithms that exploit the sparse characteristics of EEGs to classify these signals. Different feature vectors are extracted from EEG trials recorded by electrodes placed on the scalp. In this thesis, features from a small spatial region are approximated by a sparse linear combination of few atoms from a multi-class dictionary constructed from the features of the EEG training signals for each class. This is used to classify the signals based on the pattern of their sparse representation using a minimum-residual decision rule. We first attempt to use all the available electrodes to verify the effectiveness of the proposed methods. To support real time BCI, the electrodes are reduced to those near the sensorimotor cortex which are believed to be crucial for motor preparation and imagination. In a second approach, we try to incorporate the effect of spatial correlation across the neighboring electrodes near the sensorimotor cortex. To this end, instead of considering one feature vector at a time, we use a collection of feature vectors simultaneously to find the joint sparse representation of these vectors. Although we were not able to see much improvement with respect to the first approach, we envision that such improvements could be achieved using more refined models that can be subject of future works. The performance of the proposed approaches is evaluated using different features, including wavelet coefficients, energy of the signals in different frequency sub-bands, and also entropy of the signals. The results obtained from real data demonstrate that the combination of energy and entropy features enable efficient classification of motor imagery EEG trials related to hand and foot movements. This underscores the relevance of the energies and their distribution in different frequency sub-bands for classifying movement-specific EEG patterns in agreement with the existence of different levels within the alpha band. The proposed approach is also shown to outperform the state-of-the-art algorithm that uses feature vectors obtained from energies of multiple spatial projections.

This thesis discusses the development and performance of an algorithm created to calculate satellite attitude based on the comparison of satellite "physical feature" models to information derived from edge detection performed on imagery of the satellite. The quality of this imagery could range from the very clear, close-up imagery that may come from an unmanned satellite servicing mission to the faint, unclear imagery that may come from a ground-based telescope investigating a satellite anomaly. Satellite "physical feature" models describe where an edge is likely to appear in an image. These are usually defined by physical edges on the structure of the satellite or areas where there are distinct changes in material property. The theory behind this concept is discussed as well as two different approaches to implement it. Various simple examples are used to demonstrate the feasibility of the concept. These examples are well-controlled image simulations of simple physical models with known attitude. The algorithm attempts to perform the edge detection and edge registration of the simulated image and calculate the most likely attitude. Though complete autonomy was not achieved during this effort, the concept and approach show applicability.

Carried out in collaboration with the Occoquan Water Monitoring Lab, this thesis presents the results of research that sought to ascertain the spatial distribution of chlorophyll-a concentrations in the Lake Manassas Reservoir using a combination of Landsat TM satellite imagery and ground based field measurements. Images acquired on May 14, 1998 and March 8, 2000 were analyzed with chlorophyll-a measurements taken on 13, 1998 and March 7, 2000. A ratio of Landsat TM band 3: Landsat Band 4 was used in a regression with data collected at eight water quality monitoring stations run by the Occoquan Watershed Monitoring Lab. Correlation coefficients of 0.76 for the 1998 data and 0.73 for the 2000 data were achieved. Cross validation statistical analysis was used to check the accuracy of the two models. The standard error and error of the estimate were reasonable for the models from both years. In each instance, the ground data was retrieved approximately 24 hours before the Landsat Image acquisition and was a potential source of error. Other sources of error were the small sample size of chlorophyll-a concentration measurements, and the uncertainty involved in the location of the water quality sampling stations. / Master of Science

To address accuracy concerns of the National Land Cover Dataset (NLCD), this case study compares impervious surface from the NLCD to a Detailed Land Cover Dataset (DLCD) for the Town of Blacksburg, Virginia over two time periods (2005/2006 and 2011) at spatial aggregation scales (fine to coarse) and scopes (site-specific to area-extent). When comparing the total impervious surface area, the NLCD overestimated the DLCD by appreciable amounts (12-27%) for the entire town and across all specified land use zones for both time periods examined. A binary pixel-wise accuracy assessment of impervious surface revealed that the NLCD performed well for all scopes except for the single family land use zone (user accuracy <40%). The spatial aggregation of pixels to 90-m led to improved agreement between the two datasets. Using the DLCD as a reference, an empirical normalization equation was successfully applied to the NLCD to further reduce overestimation and data skewness. / Master of Science

In the current study, I examined the efficacy of mindful self-compassion, imagery, and goal-setting (i.e., treatment as usual) interventions on athletic identity, knee self-efficacy, subjective knee functioning, and perceived injustice, following ACL surgery. Twenty-nine adolescent and young adult athletes participated in the interventions and completed self-report measures assessing each of these constructs prior to their surgery and over seven weeks post-ACL surgery. HLM analyses demonstrated significant decreases in athletic identity and increases in subjective knee functioning from pre-surgery through seven weeks post-surgery. Intervention group further explained these decreases, though no one intervention clearly emerged as more or less beneficial. No significant changes were observed for athletes' ratings of knee self-efficacy or perceived injustice. Limitations and areas for future research are discussed.

This study looks at the changes to NDVI value in the Weeks Bay Reserve following the impact by Hurricane Katrina. Four Landsat images from March 24, 2005 (Pre-Katrina), September 16, 2005/ April 26, 2006 (Post-Katrina) and August 7, 2002 (Control) were classified into different landcover types and run with the NDVI vegetation index. Those images were compared against each other and showed that the September image had a NDVI value drop of 49% and the April image had a 47% drop as compared to the previous March. The emergent vegetation surrounding the shoreline was most susceptible to changes in NDVI value and recovered the slowest of the tested landcover types. Swift tracks, bay areas, and rivers in the study area where tested and showed that the rivers are the most susceptible change in NDVI value and recovered the slowest.

No description available.

No description available.