Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
- Discipline
-
- Engineering (5)
- Computer Engineering (4)
- Computer Sciences (4)
- Physics (4)
- Optics (3)
-
- Aerospace Engineering (2)
- Artificial Intelligence and Robotics (2)
- Digital Communications and Networking (2)
- Information Security (2)
- Navigation, Guidance, Control and Dynamics (2)
- Atomic, Molecular and Optical Physics (1)
- Computer and Systems Architecture (1)
- Other Computer Engineering (1)
- Probability (1)
- Robotics (1)
- Statistics and Probability (1)
- Theory and Algorithms (1)
Articles 1 - 8 of 8
Full-Text Articles in Physical Sciences and Mathematics
Optical Flow-Based Odometry For Underground Tunnel Exploration, Terra Kier
Optical Flow-Based Odometry For Underground Tunnel Exploration, Terra Kier
Theses and Dissertations
As military operations in degraded or GPS-denied environments continue to increase in frequency and importance, there is an increased necessity to be able to determine precision location within these environments. Furthermore, authorities are finding a record number of tunnels along the U.S.-Mexico border; therefore, underground tunnel characterization is becoming a high priority for U.S. Homeland Security as well. This thesis investigates the performance of a new image registration technique based on a two camera optical- flow configuration using phase correlation techniques. These techniques differ from other image based navigation methods but present a viable alternative increasing autonomy and answering the …
Context Aided Tracking With Adaptive Hyperspectral Imagery, Andrew C. Rice
Context Aided Tracking With Adaptive Hyperspectral Imagery, Andrew C. Rice
Theses and Dissertations
A methodology for the context-aided tracking of ground vehicles in remote airborne imagery is developed in which a background model is inferred from hyperspectral imagery. The materials comprising the background of a scene are remotely identified and lead to this model. Two model formation processes are developed: a manual method, and method that exploits an emerging adaptive, multiple-object-spectrometer instrument. A semi-automated background modeling approach is shown to arrive at a reasonable background model with minimal operator intervention. A novel, adaptive, and autonomous approach uses a new type of adaptive hyperspectral sensor, and converges to a 66% correct background model in …
Estimating Anthropometric Marker Locations From 3-D Ladar Point Clouds, Matthew J. Maier
Estimating Anthropometric Marker Locations From 3-D Ladar Point Clouds, Matthew J. Maier
Theses and Dissertations
An area of interest for improving the identification portion of the system is in extracting anthropometric markers from a Laser Detection and Ranging (LADAR) point cloud. Analyzing anthropometrics markers is a common means of studying how a human moves and has been shown to provide good results in determining certain demographic information about the subject. This research examines a marker extraction method utilizing principal component analysis (PCA), self-organizing maps (SOM), alpha hulls, and basic anthropometric knowledge. The performance of the extraction algorithm is tested by performing gender classification with the calculated markers.
Holistic Network Defense: Fusing Host And Network Features For Attack Classification, Jenny W. Ji
Holistic Network Defense: Fusing Host And Network Features For Attack Classification, Jenny W. Ji
Theses and Dissertations
This work presents a hybrid network-host monitoring strategy, which fuses data from both the network and the host to recognize malware infections. This work focuses on three categories: Normal, Scanning, and Infected. The network-host sensor fusion is accomplished by extracting 248 features from network traffic using the Fullstats Network Feature generator and from the host using text mining, looking at the frequency of the 500 most common strings and analyzing them as word vectors. Improvements to detection performance are made by synergistically fusing network features obtained from IP packet flows and host features, obtained from text mining port, processor, logon …
Overcoming Pose Limitations Of A Skin-Cued Histograms Of Oriented Gradients Dismount Detector Through Contextual Use Of Skin Islands And Multiple Support Vector Machines, Jonathon R. Climer
Overcoming Pose Limitations Of A Skin-Cued Histograms Of Oriented Gradients Dismount Detector Through Contextual Use Of Skin Islands And Multiple Support Vector Machines, Jonathon R. Climer
Theses and Dissertations
This thesis provides a novel visualization method to analyze the impact that articulations in dismount pose and camera aspect angle have on histograms of oriented gradients (HOG) features and eventual detections. Insights from these relationships are used to identify limitations in a state of the art skin cued HOG dismount detector's ability to detect poses not in a standard upright stances. Improvements to detector performance are made by further leveraging available skin information, reducing false detections by an additional order of magnitude. In addition, a method is outlined for training supplemental support vector machines (SVMs) from computer generated data, for …
Kernelized Locality-Sensitive Hashing For Fast Image Landmark Association, Mark A. Weems
Kernelized Locality-Sensitive Hashing For Fast Image Landmark Association, Mark A. Weems
Theses and Dissertations
As the concept of war has evolved, navigation in urban environments where GPS may be degraded is increasingly becoming more important. Two existing solutions are vision-aided navigation and vision-based Simultaneous Localization and Mapping (SLAM). The problem, however, is that vision-based navigation techniques can require excessive amounts of memory and increased computational complexity resulting in a decrease in speed. This research focuses on techniques to improve such issues by speeding up and optimizing the data association process in vision-based SLAM. Specifically, this work studies the current methods that algorithms use to associate a current robot pose to that of one previously …
A Multi Agent System For Flow-Based Intrusion Detection Using Reputation And Evolutionary Computation, David Hancock
A Multi Agent System For Flow-Based Intrusion Detection Using Reputation And Evolutionary Computation, David Hancock
Theses and Dissertations
The rising sophistication of cyber threats as well as the improvement of physical computer network properties present increasing challenges to contemporary Intrusion Detection (ID) techniques. To respond to these challenges, a multi agent system (MAS) coupled with flow-based ID techniques may effectively complement traditional ID systems. This paper develops: 1) a scalable software architecture for a new, self-organized, multi agent, flow-based ID system; and 2) a network simulation environment suitable for evaluating implementations of this MAS architecture and for other research purposes. Self-organization is achieved via 1) a reputation system that influences agent mobility in the search for effective vantage …
A Multispectral Bidirectional Reflectance Distribution Function Study Of Human Skin For Improved Dismount Detection, Bradley M. Koch
A Multispectral Bidirectional Reflectance Distribution Function Study Of Human Skin For Improved Dismount Detection, Bradley M. Koch
Theses and Dissertations
In 2008, the Sensors Exploitation Research Group at the Air Force Institute of Technology began using spectral properties of skin for the detection and classification of humans. Since then a multispectral skin detection system was developed to exploit the optical properties of human skin at wavelengths in the visible and near infrared region of the electromagnetic spectrum. A rules-based detector, analyzing an image spectrally, currently bases its skin pixel selection criteria on a diffuse skin reflectance model. However, when observing skin in direct view of the sun, a glint of light off skin is common and indicates specularity. The areas …