Papers by Terry Huntsberger
Neural Systems for Motion Analysis: Single Neuron and Network Approaches
Studies in fuzziness and soft computing, 2000
ABSTRACT . 1 Introduction Early research in computer vision was concerned mainly with static imag... more ABSTRACT . 1 Introduction Early research in computer vision was concerned mainly with static images. Our previous work concentrated on analysis of static images for segmentation purposes [1,24]. The method used a fuzzy self-organizing feature map (FSOFM) [31] as part of a multistage neural network modeled after the rattlesnake optic tectum for the fusion of two sensory modalities [1]. Results comparable to and better than traditional evidence combination schemes were obtained [24]. For some applications, the analysis of static images may suer from inherent limitations in the amount of three{dimensional information that can be extracted. Recently, dynamic scene analysis in the machine vision eld has been given increasing attention [4,37,56] (and references therein). The input to a dynamic scene analysis system is a series of images, each representing the scene at a particular time instant. The images can be produced by disparate sensors, giving information about each time step in multiple c...
Journal of Petroleum Geology, Oct 1, 1987
Three methods are presented for determining paleo heat-flux from fission scar tracks in detrital ... more Three methods are presented for determining paleo heat-flux from fission scar tracks in detrital apatite collected from wells. The annealing of thejission scar tracks is modelled using timetemperature integrals assuming first-order kinetics in agreement with the laboratory experiments over the last f$een years. The three methods invert the present-day fission-track well data along the burial paths of the sedimentary units. Thejirst two methods involve inverting the depth-dependent (a) areal track-number density pergm of uranium and (b) mean track length to obtain an estimate of the temporal dependence of the paleo heat-flux Since these methods tend to overlook short tracks (method (a)), and be biased towards longer tracks (method @) ) , they tend toprovidepaleo heat-flux variations which are too high and too low, respectively. The uncertainty in the determination ofpaleo heat-flux variation is also quite high, as is shown by using data @om two wells: Kambara-1 and Cumnga-1 in the NW Canning Basin, Australia. The third method inverts all of the individual distributions of track lengths at each sampled depth in a well. The sensitivity of this method is much higher than methods (a) or (b). In fact, we are able to determine not only the rate of change of heat-flux with time but, for thejirst time, variations in the direction of change of heat-flw allowing us to quantify heating and cooling events, as opposed to just heating or cooling events. In addition, we are able to determine the rate constant for track annealing and, in line with qualitative arguments advanced by Gleadow et a l (1983), we find that the rate constant appropriate to the multi-million year time-scale of subsurfce apatite track annealing is lower, by a factor of about three, than the value advocated from short-lived (on a geologic time-scale) laboratory experiments. Detailed track data from two wells, Kambara-1 and Curringa-I, yield essentially identical values for the rate constant, but very dixerent thermal history behaviours.
Hypercube algorithms suitable for image understanding in uncertain environments
Parallel algorithms for color image segmentation and edge detection are presented and analyzed. T... more Parallel algorithms for color image segmentation and edge detection are presented and analyzed. These algorithms are part of a recently developed computer vision system which uses multiple-valued logic to represent uncertainty in the imaging process and in intermediate results. Algorithms for the extraction of three-dimensional properties of objects using dynamic scene analysis techniques within the same computer vision system framework
<title>Edge dipole and edge field for boundary detection</title>
Proceedings of SPIE, Jul 8, 1998
ABSTRACT Conventionally, edges are treated as either scalar or vector quantities. This paper pres... more ABSTRACT Conventionally, edges are treated as either scalar or vector quantities. This paper presents a novel framework which treats edges as directional dipoles that induce the field around themselves. An analogy can be made between this concept and the interaction of magnetic dipoles with the magnetic field. The dipoles interact with the field and align themselves into a smooth contour configuration. This paper shows the effectiveness of the concept in edge linking and proposes efficient computational schemes for real-time implementation of the edge dipole interactions. It also proposes an image representation using the dipoles on a hexagonal lattice and a contour extraction algorithm implemented on the representation. The algorithm consists of three processes: noise removal, edge alignment and edge thinning/extension. The results of some experimental studies are also presented. 1. INTRODUCTION An edge is an important cue for the human vision processing and has been an important feature for...
Lecture Notes in Computer Science, 2001
The paper describes a robust edge and contour extraction technique under two types of degradation... more The paper describes a robust edge and contour extraction technique under two types of degradation: random noise and aliasing. The technique employs unambiguous probabilistic relaxation to distinguish features from noise and refine their spatial locations at sub-pixel accuracy. The most important component in the probabilistic relaxation is a compatibility function. The paper suggests a function with which the optimal orientation of edges can be derived analytically, thus allowing an efficient implementation of the relaxation process. A contour extraction algorithm is designed by combining the relaxation process and a perceptual organization technique. Results on both synthetic and natural images are given and show effectiveness of our approach against noise and aliasing.
Since the form of the nuclear force is not explicitly known, various models have been proposed to... more Since the form of the nuclear force is not explicitly known, various models have been proposed to account for the nucleon-nucleon scattering data and gross properties of nuclei, such as quadrupole moments, binding energy, etc. Some of the better known models are the liquid-drop [7], shell [4], unified [5] and grouptheoretical [6] approximations. This paper presents a hypercube based algorithm for the derivation of nuclear structure energy levels and binding energies for a wide range o isotopes. The algorithm is based on previous work f l , 21 which studied isotopes in the f-p-g shell shown in Figure . Extensions to higher order shells are obtained using a distributed decomposition of the coupled state and perturbation calculations. The results of scaling experiments for the NCUBE/lO system at the University of South Carolina are also included.
Proceedings of the International Conference on Automated Planning and Scheduling, May 25, 2021
We propose a new method to analyze the temporal brittleness of task networks, which allows the de... more We propose a new method to analyze the temporal brittleness of task networks, which allows the detection and enumeration of activities that, with modest task execution duration variation make the execution of the task network dynamically uncontrollable. In this method, we introduce a metric for measuring an activity brittleness -defined as the degree of acceptable deviation of its nominal duration -and describe how that measurement is mapped to task network structure. Complementary to existing work on plan robustness analysis which informs how likely a task network is to succeed or not, the proposed analysis and metric go deeper to pinpoint the sources of potential brittleness due to temporal constraints and to focus either human designers and/or automated task network generators (e.g. scheduler/planners) to address sources of undesirable brittleness. We apply the approach to a set of task networks (called sol types) in development for NASA's next planetary rover and present common patterns that are sources of brittleness. These techniques are currently under evaluation for potential use supporting operations of the Mars 2020 rover.
Autonomous Target Tracking for Asteroid Landing
Asteroid landing of spacecraft using autonomous target tracking based on passive electro-optical ... more Asteroid landing of spacecraft using autonomous target tracking based on passive electro-optical sensing information is presented. The Gabor filters are used for the recognition and tracking of planetary features. A new guidance and navigation method based on matching image features based on stereo cameras is proposed. Simulation testing has been conducted to verify the passive guidance approach.
Self-organizing neural networks for unsupervised color image recognition
Presents a new self-organizing neural network system for color image recognition for any given im... more Presents a new self-organizing neural network system for color image recognition for any given image data set without a priori information about the number of clusters or cluster centers. The system has a self-organizing feature that utilizes multiple valued information in the process of updating weights between the input layer and distance layer. This model has the shape of a one dimensional ring-structure, with every neuron influencing its two nearest neighbors. Input vectors are distributed to each neuron in parallel. The model showed good convergence properties for several test data sets. Comparisons with original color images and reconstructed images are also presented.<<ETX>>
Optical Engineering, Mar 1, 1998
Robust pattern recognition within the Bayesian framework for scene segmentation/boundary detectio... more Robust pattern recognition within the Bayesian framework for scene segmentation/boundary detection is oftentimes hampered by the presence of textures within natural images. In order to improve segmentation/boundary detection on natural images, it is necessary to combine multiple features e ectively. This paper introduces two algorithms for combining both color and texture features to assist boundary detection processes. One is to combine features through the surface processes and the other through the line processes. The algorithms can be generalized for combining any number of feature sets.
Command of support robots by the warfighter requires intuitive interfaces to quickly communicate ... more Command of support robots by the warfighter requires intuitive interfaces to quickly communicate high degree-offreedom (DOF) information while leaving the hands unencumbered. Stealth operations rule out voice commands and vision-based gesture interpretation techniques, as they often entail silent operations at night or in other low visibility conditions. Targeted at using bio-signal inputs to set navigation and manipulation goals for the robot (say, simply by pointing), we developed a system based on an electromyography (EMG) "BioSleeve", a high density sensor array for robust, practical signal collection from forearm muscles. The EMG sensor array data is fused with inertial measurement unit (IMU) data. This paper describes the BioSleeve system and presents initial results of decoding robot commands from the EMG and IMU data using a BioSleeve prototype with up to sixteen bipolar surface EMG sensors. The BioSleeve is demonstrated on the recognition of static hand positions (e.g. palm facing front, fingers upwards) and on dynamic gestures (e.g. hand wave). In preliminary experiments, over 90% correct recognition was achieved on five static and nine dynamic gestures. We use the BioSleeve to control a team of five LANdroid robots in individual and group/squad behaviors. We define a gesture composition mechanism that allows the specification of complex robot behaviors with only a small vocabulary of gestures/commands, and we illustrate it with a set of complex orders.
Cognitive architecture for mixed human-machine team interactions for space exploration
... silhouettes. The optical flow is extracted using the Black and Anandan algorithm [44] that an... more ... silhouettes. The optical flow is extracted using the Black and Anandan algorithm [44] that analyzes two frames at a time to solve the optical flow equation within a pyramid hierarchy of lower resolution versions of the two frames. ...
Proceedings of SPIE, Oct 9, 1998
This paper describes the development o f a d v anced rover navigation and manipulation techniques... more This paper describes the development o f a d v anced rover navigation and manipulation techniques for use by NASA's Sample Return Rover. These techniques include an algorithm for estimating the change in the rover's position and orientation by registering successive range maps from the rover's hazard avoidance stereo camera pair and the fusion of this information with the rover's wheel odometry. This map registration technique is also extended to register range maps to an a priori model-based range map for relative r o ver position and orientation determination. Finally, a technique for the robust and precise positioning of a r o ver-mounted manipulator using visual feedback from the rover's stereo pair is presented. Experimental results for each of these techniques is documented in this paper.
Color edge detection
Pattern Recognition Letters, May 1, 1985
... given in (3) was applied to a full color natural scene of a house shown in Figure 3. Color sp... more ... given in (3) was applied to a full color natural scene of a house shown in Figure 3. Color spaces chosen for the experiments were the RGB space, a color space derived from the results of the Karhunen-Loeve transformation by Ohta (1980), and the CIE XYZ color space. ...
Parallel Self-Organizing Feature Maps for Unsupervised Pattern Recognition
International Journal of General Systems, May 1, 1990
... TERRANCE L. HUNTSBERGER and PONGSAK AJJIMARANGSEE Intelligent Systems Laboratory, Department ... more ... TERRANCE L. HUNTSBERGER and PONGSAK AJJIMARANGSEE Intelligent Systems Laboratory, Department of Computer Science, University of South Carolina, Columbia, South Carolina ... which classifies pixels in this way is called a hard c partition of X. When the range of Ui ...
Uploads
Papers by Terry Huntsberger