Search

The recent uptick in senseless shootings in otherwise quiet and relatively safe environments is powerful evidence of the need, now more than ever, to reduce these occurrences. Artificial intelligence (AI) can play a significant role in deterring individuals from attempting these acts of violence. The installation of audio sensors can assist in the proper surveillance of surroundings linked to public safety, which is the first step toward AI-driven surveillance. With the increasing popularity of machine learning (ML) processes, systems are being developed and optimized to assist personnel in highly dangerous situations. In addition to saving innocent lives, supporting the capture of the responsible criminals is part of the AI algorithm that can be hosted in acoustic gunshot detection systems (AGDSs). Although there has been some speculation that these AGDSs produce a higher false positive rate (FPR) than reported in their specifications, optimizing the dataset used for the model’s training and testing will enhance its performance. This dissertation proposes a new gunshot-like sound database that can be incorporated into a dataset for improved training and testing of a ML gunshot detection model. Reduction of the sample bias (that is, a bias in ML caused by an incomplete database) is achievable. The Mel frequency cepstral coefficient (MFCC) feature extraction process was utilized in this research. The uniform manifold and projection (UMAP) algorithm revealed that the MFCCs of this newly created database were the closest sounds to a gunshot sound, as compared to other gunshot-like sounds reported in literature. The UMAP algorithm reinforced the outcome derived from the calculation of the distances of the centroids of various gunshot-like sounds in MFCCs’ clusters. Further research was conducted into the feature reduction aspect of the gunshot detection ML model. Reducing a feature set to a minimum, while also maintaining a high accuracy rate, is a key parameter of a hi, 2022, Includes bibliography., Degree granted: Dissertation (PhD)--Florida Atlantic University, 2022., Collection: FAU Electronic Theses and Dissertations Collection

Summary: In the last decade, deep learning models have been successfully applied to a variety of applications and solved many tasks. The ultimate goal of this study is to produce deep learning models to improve the skills of forecasting ocean dynamic events in general and those of the Loop Current (LC) system in particular. A specific forecast target is to predict the geographic location of the (LC) extension and duration, LC eddy shedding events for a long lead time with high accuracy. Also, this study aims to improve the predictability of velocity fields (or more precisely, velocity volumes) of subsurface currents. In this dissertation, several deep learning based prediction models have been proposed. The core of these models is the Long-Short Term Memory (LSTM) network. This type of recurrent neural network is trained with Sea Surface Height (SSH) and LC velocity datasets. The hyperparameters of these models are tuned according to each model's characteristics and data complexity. Prior to training, SSH and velocity data are decomposed into their temporal and spatial counterparts.A model uses the Robust Principle Component Analysis is first proposed, which produces a six-week lead time in forecasting SSH evolution. Next, the Wavelet+EOF+LSTM (WELL) model is proposed to improve the forecasting capability of a prediction model. This model is tested on the prediction of two LC eddies, namely eddy Cameron and Darwin. It is shown that the WELL model can predict the separation of both eddies 10 and 14 weeks ahead respectively, which is two more weeks than the DAC model. Furthermore, the WELL model overcomes the problem due to the partitioning step involved in the DAC model. For subsurface currents forecasting, a layer partitioning method is proposed to predict the subsurface field of the LC system. A weighted average fusion is used to improve the consistency of the predicted layers of the 3D subsurface velocity field. The main challenge of forecasting of the LC and its, 2021, Includes bibliography., Degree granted: Dissertation (PhD)--Florida Atlantic University, 2021., Collection: FAU Electronic Theses and Dissertations Collection

Deep learning models have been successfully applied to a variety of machine learning tasks, including image identification, image segmentation, object detection, speaker recognition, natural language processing, bioinformatics and drug discovery, among other things. This dissertation introduces Multi-Model Deep Learning (MMDL), a new ensemble deep learning approach for signal classification and event forecasting. The ultimate goal of the MMDL method is to improve classification and forecasting performances of individual classifiers by fusing results of participating deep learning models. The performance of such an ensemble model, however, depends heavily on the following two design features. Firstly, the diversity of the participating (or base) deep learning models is crucial. If all base deep learning models produce similar classification results, then combining these results will not provide much improvement. Thus, diversity is considered to be a key design feature of any successful MMDL system. Secondly, the selection of a fusion function, namely, a suitable function to integrate the results of all the base models, is important. In short, building an effective MMDL system is a complex and challenging process which requires deep knowledge of the problem context and a well-defined prediction process. The proposed MMDL method utilizes a bank of Convolutional Neural Networks (CNNs) and Stacked AutoEncoders (SAEs). To reduce the design complexity, a randomized generation process is applied to assign values to hyperparameters of base models. To speed up the training process, new feature extraction procedures which captures time-spatial characteristics of input signals are also explored. The effectiveness of the MMDL method is validated in this dissertation study with three real-world case studies. In the first case study, the MMDL model is applied to classify call types of groupers, an important fishery resource in the Caribbean that produces sounds associated with rep, 2019, Includes bibliography., Degree granted: Dissertation (Ph.D.)--Florida Atlantic University, 2019., Collection: FAU Electronic Theses and Dissertations Collection

Reducing the amount of radiation in X-ray computed tomography has been an active area of research in the recent years. The reduction of radiation has the downside of degrading the quality of the CT scans by increasing the ratio of the noise. Therefore, some techniques must be utilized to enhance the quality of images. In this research, we approach the denoising problem using two class of algorithms and we reduce the noise in CT scans that have been acquired with 75% less dose to the patient compared to the normal dose scans. Initially, we implemented wavelet denoising to successfully reduce the noise in low-dose X-ray computed tomography (CT) images. The denoising was improved by finding the optimal threshold value instead of a non-optimal selected value. The mean structural similarity (MSSIM) index was used as the objective function for the optimization. The denoising performance of combinations of wavelet families, wavelet orders, decomposition levels, and thresholding methods were investigated. Results of this study have revealed the best combinations of wavelet orders and decomposition levels for low dose CT denoising. In addition, a new shrinkage function is proposed that provides better denoising results compared to the traditional ones without requiring a selected parameter. Alternatively, convolutional neural networks were employed using different architectures to resolve the same denoising problem. This new approach improved denoising even more in comparison to the wavelet denoising., 2018, Includes bibliography., Degree granted: Dissertation (Ph.D.)--Florida Atlantic University, 2018., Collection: FAU Electronic Theses and Dissertations Collection

Sickle cell disease is an inherited blood cell disorder that affects about 100,000 people in the US and results in high cost of medical care exceeding $1.1 billion annually. Sickle cell patients suffer from unpredictable, painful vaso-occlusive crises. Portable, costeffective approaches for diagnosis and monitoring sickle blood activities are important for a better management of the disease and reducing the medical cost. In this research, a mobile application controlled, impedance-based flow cytometer is developed for the diagnosis of sickle cell disease. Calibration of the portable device is performed using a component of known impedance value. The preliminary test results are then compared to those obtained by a commercial benchtop impedance analyzer for further validation. With the developed portable flow cytometer, experiments are performed on two sickle cell samples and a healthy cell sample. The acquired results are subsequently analyzed with MATLAB scripts to extract single-cell level impedance information as well as statistics of different cell conditions. Significant differences in cell impedance signals are observed between sickle cells and normal cells, as well as between sickle cells under hypoxia and normoxia conditions., 2018, Includes bibliography., Degree granted: Thesis (M.S.)--Florida Atlantic University, 2018., Collection: FAU Electronic Theses and Dissertations Collection

Summary: Ocean is home to a large population of marine mammals such as dolphins and whales and concerns over anthropogenic activities in the regions close to their habitants have been increased. Therefore the ability to detect the presence of these species in the field, to analyze and classify their vocalization patterns for signs of distress and distortion of their communication calls will prove to be invaluable in protecting these species. The objective of this research is to investigate methods that automatically detect and classify vocalization patterns of marine mammals. The first work performed is the classification of bottlenose dolphin calls by type. The extraction of salient and distinguishing features from recordings is a major part of this endeavor. To this end, two strategies are evaluated with real datasets provided by Woods Hole Oceanographic Institution: The first strategy is to use contour-based features such as Time-Frequency Parameters and Fourier Descriptors and the second is to employ texture-based features such as Local Binary Patterns (LBP) and Gabor Wavelets. Once dolphin whistle features are extracted for spectrograms, selection of classification procedures is crucial to the success of the process. For this purpose, the performances of classifiers such as K-Nearest Neighbor, Support Vector Machine, and Sparse Representation Classifier (SRC) are assessed thoroughly, together with those of the underlined feature extractors., 2014, Includes bibliography., Degree granted: Dissertation (Ph.D.)--Florida Atlantic University, 2014., Collection: FAU Electronic Theses and Dissertations Collection

Summary: Research presented in this thesis develops a mainly theoretical basis and computer models for enhancing the throughput of multi-user wireless communication networks. The cross-layer combination of an adaptive modulation and coding (AMC) scheme at the physical layer and the use of automatic repeat request (ARQ) retransmi ssions at the data link layer is integrated into a scheduling framework for multi-user networks. Scheduling algorithms incorporating retransmission diversity are derived for three cases of typical network traffic: best-effort, non-realtime, and realtime. For each case, numeric computer si mulations of wireless communications over Nakagami-m block fading channels are developed to examine the effectiveness of the formulated schemes., 2008, College of Engineering and Computer Science, Degree granted: Thesis (M.S.)--Florida Atlantic University, 2008., Collection: FAU Electronic Theses and Dissertations Collection

The gripping action as performed by an average person is developed over their life and changes over time. The initial learning is based on trial and error and becomes a natural action which is modified as the physiology of the individual changes. Each grip type is a personal expression and as the grip changes over time to accommodate physiologically changes, it can be considered to be a grip-signature. lt is postulated that an ANN can deliver a classification mechanism that is able to make sense of the varying gripping inputs that are linearly inseparable and uniquely attributed to user physiology. Succinctly, in this design, the stifnulus is characterized by a voltage that represents the applied force in a grip. This signature of forces is then used to train an ANN to recognize the grip that produced the signature, the ANN in turn is used to successfully classify three unique states of grip-signatures collected from the gripping action of various individuals as they hold, lift and crush a paper coffee-cup. A comparative study is done for three types of classification: K-Means, Backpropagation Feedforward Neural Networks and Recurrent Neural Networks, with recommendations made in selecting more effective classification methods., 2007, College of Engineering and Computer Science, Degree granted: Thesis (M.S.)--Florida Atlantic University, 2007., Collection: FAU Electronic Theses and Dissertations Collection

Autonomous video surveillance systems are usually built with several functional blocks such as motion detection, foreground and background separation, object tracking, depth estimation, feature extraction and behavioral analysis of tracked objects. Each of those blocks is usually designed with different techniques and algorithms, which may need significant computational and hardware resources. In this thesis we present a surveillance system based on an optical flow concept, as a main unit on which other functional blocks depend. Optical flow limitations, capabilities and possible problem solutions are discussed in this thesis. Moreover, performance evaluation of various methods in handling occlusions, rigid and non-rigid object classification, segmentation and tracking is provided for a variety of video sequences under different ambient conditions. Finally, processing time is measured with software that shows an optical flow hardware block can improve system performance and increase scalability while reducing the processing time by more than fifty percent., 2007, College of Engineering and Computer Science, Degree granted: Thesis (M.S.)--Florida Atlantic University, 2007., Collection: FAU Electronic Theses and Dissertations Collection

Summary: A method to create 3D-face image using 2D-face images is the objective of this research. The 3D-face image is constructed using a set of 3D-face images of other persons available in a face database. The 3D-face image actually depicts a parameterized form in terms of depth and texture. This concept can be used to facilitate creating a 3D-face image from 2D database. For this purpose, a 3D-face database is first developed. When a 2D-face image is presented to the system, a 3D-face image that starts with an average 3D-face image (derived from the 3D-face database) is projected onto the 2D-image plane, with necessary rotation, translation, scaling and interpolation. The projected image is then compared with the input image; and, an optimization algorithm is applied to minimize an error index by selecting 3D-depth and texture parameters. Hence, the projected image is derived. Once the algorithm converges, the resulting 3D-depth and the texture parameters can be employed to construct a 3D-face image of the subject photographed in the 2D-images. A merit of this method is that only the depth and texture parameters of the compared images are required to be stored in the database. Such data can be used either for the recreation of a 3D-image of the test subject or for any biometric authentication (based on 3D face recognition). Results from an experimental study presented in the thesis illustrate the effectiveness of the proposed approach, which has applications in biometric authentication and 3D computer graphics areas., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 2006.

Summary: Design and Tuning a fuzzy logic controller (FLCs) are usually done in two stages. In the first stage, the structure of a FLC is determined based on physical characteristics of the system. In the second stage, the parameters of the FLC are selected to optimize the performance of the system. The task of tuning FLCs can be performed by a number of methods such as adjusting control gains, changing membership functions, modifying control rules and varying control surfaces. A method for the design and tuning of FLCs through modifying their control surfaces is presented in this dissertation. The method can be summarized as follows. First, fuzzy control surfaces are modeled with Bezier functions. Shapes of the control surface are then adjusted through varying Bezier parameters. A Genetic Algorithm (GA) is used to search for the optimal set of parameters based on the control performance criteria. Then, tuned control surfaces are sampled to create rule-based FLCs. To further improve the system performance, continuity constraints of the curves are imposed. Under the continuity constraints with the same number of tunable parameters, one can obtain more flexible curves that have the potential to improve the overall system performance. An important issue is to develop a new method to self-tune a fuzzy PD controller. The method is based on two building blocks: (I) Bezier functions used to model the control surfaces of the fuzzy PD controller; and, shapes of control surfaces are then adjusted by varying Bezier parameters. (II) The next step involves using a gradient-based optimization algorithm with which the input scaling factors and Bezier parameters are on-line tuned until the controller drives the output of the process as close as possible to the reference position. To protect vendors and consumers from being victimized, various trust models have been used in e-commerce practices. However, a strict verification and authentication process may pose unnecessary heavy cost, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 2005.

Summary: In certain applications, one needs to control physical plants that operate in hazardous conditions. In such situations, it is necessary to acquire access to the controller from a different (remote) location through data communication networks, in order to interconnect the remote location and the controller. The use of such network linking between the plant and the controller may introduce network delays, which would affect adversely the performance of the process control. The main theoretical contribution of this thesis is to answer the following question: How large can a network delay be tolerated such that the delayed closed-loop system is locally asymptotically stable? An explicit time-independent bound for the delay is derived. In addition, various practical realizations for the remote control tasks are presented, utilizing a set of predefined classes for serial communication, data-acquisition modules and stream-based sockets. Due to the presence of a network, implementing an efficient control scheme is a not trivial problem. Hence, two practical frameworks for Internet-based control are illustrated in this thesis. Related implementation issues are addressed in detail. Examples and case studies are provided to demonstrate the effectiveness of the proposal approach., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 2003.

Summary: The design of any communication receiver needs to addresses the issues of operating under the lowest possible signal-to-noise ratio. Among various algorithms that facilitate this objective are those used for iterative decoding of two-dimensional systematic convolutional codes in applications such as spread spectrum communications and Code Division Multiple Access (CDMA) detection. A main theme of any decoding schemes is to approach the Shannon limit in signal-to-noise ratio. All these decoding algorithms have various complexity levels and processing delay issues. Hence, the optimality depends on how they are used in the system. The technique used in various decoding algorithms is termed as iterative decoding. Iterative decoding was first developed as a practical means for decoding turbo codes. With the Log-Likelihood algebra, it is shown that a decoder can be developed that accepts soft inputs as a priori information and delivers soft outputs consisting of channel information, a posteriori information and extrinsic information to subsequent stages of iteration. Different algorithms such as Soft Output Viterbi Algorithm (SOVA), Maximum A Posteriori (MAP), and Log-MAP are compared and their complexities are analyzed in this thesis. A turbo decoder is implemented on the Digital Signal Processing (DSP) chip, TMS320C30 by Texas Instruments using a Modified-Log-MAP algorithm. For the Modified-Log-MAP-Algorithm, the optimal choice of the lookup table (LUT) is analyzed by experimenting with different LUT approximations. A low complexity decoder is proposed for a (7,5) code and implemented in the DSP chip. Performance of the decoder is verified under the Additive Wide Gaussian Noise (AWGN) environment. Hardware issues such as memory requirements and processing time are addressed for the chosen decoding scheme. Test results of the bit error rate (BER) performance are presented for a fixed number of frames and iterations., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 2002.

Summary: In the last 10 years, due to the rapid developments in computers and Internet, the Electronic Commerce has advanced significantly. More and more companies have shifted their businesses activities to the Internet. However, the popular use of ecommerce has also raised serious security problems. Therefore developing security schemes has become a key issue both in the academic as well as industrial research. Since the Internet is open to the public, the associated security issue is challenging. A good security strategy should not only protect the vendors' interest, but also enhance the mutual trust between vendors and customers. As a result, the people will feel more confident in conducting e-commerce. This thesis is dedicated to develop a fuzzy-logic based trust model. In general, the ecommerce transactions need costly verification and authentication process. In some cases, it is not cost effective to verify and authenticate each transaction, especially for transactions involving only small amount of money and for customers having an excellent transaction history. In view of this, in this research a model that distinguishes potentially safe transactions from unsafe transactions is developed. Only those potentially unsafe transactions need to be verified and authenticated. The model takes a number of fuzzy variables as inputs. However, this poses problems in constructing the trust table since the number of fuzzy rules will increase exponentially as the number of fuzzy variables increase. To make the problem more trackable, the variables are divided into several groups, two for each table. Each table will produce a decision on trust. The final decision is made based on the "intersection" of all these outputs. Simulation studies have been conducted to validate the effectiveness of the proposed trust model. Therefore simulations, however, need to be tested in a real business environment using real data. Relevant limitations on the proposed model are hence discusse, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 2001.

Summary: To assess and evaluate the performance of robots and machine tools dynamically, it is desirable to have a precision measuring device that performs dynamic measurement of end-effector positions of such robots and machine tools. Among possible measurement techniques, Laser Tracking Systems (LTSs) exlnbit the capability of high accuracy, large workspace, high sampling rate, and automatic target-tracking,. and thus are well-suited for robot calibration both kinematically and dynamically. In this dissertation, the design and implementation of a control system for a homemade laser tracking measurement systems is addressed and calibration of a robot using the laser tracking system is demonstrated Design and development of a control system for a LTS is a challenging task. It involves a deep understanding of laser interferometry,. controls, mechanics and optics,. both in theoretical perspective and in implementation aspect. One of the most important requirements for a successful design and implementation of a control system for the LTS is proper installation and alignment of the laser and optical system,. or laser transducer system. The precision of measurement using the LTS depends highly on the accuracy of the laser transducer system, as well as the accuracy of the installation and alignment of the optical system. Hence, in reference to the experimental alignment method presented in this dissertation, major error sources affecting the system measurement accuracy are identified and analyzed. A manual compensation method is developed to eliminate the effects of these error sources effectively in the measurement system. Considerations on proper design and installation of laser and optical components are indicated in this dissertation. As a part of the conventional control system design, a dynamic system model of the LTS is required. In this study, a detailed derivation and analysis of the dynamic model of the motor gimbal system using Lagrange-Euler equations of moti, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 2000.

Summary: Robot calibration is a software-based accuracy enhancement process. It is normally implemented in a well-controlled environment. However, for a system that function in a natural environment, it is desirable that the system is capable of performing a calibration task without any external expensive calibration apparatus and elaborate setups, i.e., system self-calibration. Vision systems have become standard automation components as cameras are normally integral components of most robotic manipulators. This research focuses on camera-aided robot self-calibration. Unlike classical vision-based robot calibration methods, which need both image coordinates and precise 3D world coordinates of calibration points, the self-calibration algorithms proposed in the dissertation only require a sequence of images of objects in a natural environment and a known scale. A new robot self-calibration algorithm using a known scale at every camera pose is proposed in the dissertation. It has been known that, the extrinsic parameters of the camera along with its intrinsic parameters can be obtained up to a scale factor by using the corresponding image points of objects due to the factor that the system is inherently under-determined. Now, if the camera is treated as the tool of the robot, one is then able to compute the corresponding robot pose directly from the camera, extrinsic parameters once the scale factor is available. This scale factor, which changes from one camera pose to another, can be uniquely determined from the known scale at each robot pose. The limitation of the above approach for robot self-calibration is that the known scale has to be utilized at every robot measurement pose. A new algorithm is proposed by using the known scale only once in the entire self-calibration procedure. The prerequisite of this calibration algorithm is a carefully planned optimal measurement trajectory for the estimation of the scale factor. By taking into consideration of the observabi, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 2000.

Summary: Laser tracking coordinate measuring machines have the potential of continuously measuring three dimensional target coordinates in a large workspace with a fast sampling rate and high accuracy. Proper calibration of a laser tracking measurement system is essential prior to use of such a device for metrology. In the absence of a more accurate instrument for system calibration, one has to rely on self-calibration strategies. In this dissertation, a kinematic model that describes not only the motion but also geometric variations of a multiple-beam laser tracking system was developed. The proposed model has the following features: (1) Target positions can be computed from both distance and angular measurements. (2) Through error analysis it was proven that even rough angular measurement may improve the overall system calibration results. A self-calibration method was proposed to calibrate intelligent machines with planar constraints. The method is also applied to the self-calibration of the laser tracking system and a standard PUMA 560 robot. Various calibration strategies utilizing planar constraints were explored to deal with different system setups. For each calibration strategy, issues about the error parameter estimation of the system were investigated to find out under which conditions these parameters can be uniquely estimated. These conditions revealed the applicability of the planar constraints to the system self-calibration. The observability conditions can serve as a guideline for the experimental setup when planar constraint is utilized in the machine calibration including the calibration of the laser tracking systems. Intensive simulation studies were conducted to check validity of the theoretical results. Realistic noise values were injected to the system models to statistically assess the behavior of the self-calibration system under real-world conditions. Various practical calibration issues were also explored in the simulations and therefore to, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1999.

Summary: With the strikingly fast development of industrial applications and research projects, control systems have become more and more complex than ever. Intelligent control techniques, featuring their being more robust and their availability when system mathematical models are unknown, have proven to be one of the most attractive and highlighted areas in the automatic control arena. This thesis concentrates first on the design of a laser tracking system. A standard design procedure of Fuzzy Logic Controllers (FLCs) is followed, which is then realized in a PC-based environment in the design. An essential issue in this thesis study is the auto tuning of the Fuzzy Logic Controller. An efficient tuning method, mu-law functions, which can adjust both the shape and scaling gain of fuzzy controller's decision table is adopted. Also a search process called Downhill Simplex Search is chosen. Combining these two methods, a Simplex-mu-law auto-tuning algorithm that fits our application is applied to tune the FLC for the laser tracking system. Another issue covered in this research is to modify the Fuzzy Logic Controller structure by changing the distribution of the membership functions. Based on the analysis of the real time error histogram of the system, a novel method is proposed in the thesis for the modification of the membership functions To assess the effectiveness of the methods proposed in this thesis, a prototype laser tracking system is constructed at the FAU Robotics Center. The control strategy proposed in this thesis is tested extensively by simulations and experimentations on the prototype system., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 1999.

Summary: To assess and improve the accuracy of an intelligent machine such as a precision robot and a computer numerically controlled machine tool, it is exceedingly desirable to have a high performance Coordinate Measuring Machine (CMM). Among various coordinate measuring devices, a laser tracking CMM has the advantages of noninvasiveness and extremely high precision over a large workspace. In this dissertation, we concentrate on the design and control of a new type of spherical gimbal for laser tracking system, whose motion is constrained by two spherical surfaces and whose axes are motorized. By this design, principal errors of a conventional tracking gimbal are reduced. To be able to integrate the laser tracking unit into an intelligent machine, a compact optical head is also designed. This laser tracking system is thus capable of being either a stand-alone or an on-line measuring device. An important issue in developing a laser tracking CMM is control. An intelligent control scheme is reported in this dissertation. The controller has the following elements: The entire tracking process of the system is classified into three modes: normal tracking, motionless and change of directions. An artificial neural network is designed to classify on-line which mode the system is in. A Fuzzy Logic Controller (FLC) suitable for the particular tracking mode is then activated to control the system. To deal with the situation in which a target suddenly changes its direction, a feed-forward compensation component is designed. Decoupling units are also added to the control scheme, by which the entire process of tracking controller design can be greatly simplified. To further improve the system performance, various structures of FLCs are analyzed in the dissertation. It is discovered that there is a constraint in the cascade proportional-integral-derivative (PID)-type FLC. Whenever this constraint is violated, the design of the controller will not be optimal. To solve this problem, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1998.

Summary: Self-calibration is a desirable feature for an intelligent machine such as a robot that must function outside of controlled laboratory conditions. This is because it is inevitable that variations in the kinematic model arise from imperfections in the manufacturing process and changes of environment conditions. Self-calibration has the potential of (a) removing the dependence on external pose sensing, (b) producing high accuracy measurement data over the entire workspace of the system with an extremely fast measurement rate, (c) being automated and completely non invasive, (d) facilitating on-line accuracy compensation, and (e) being cost effective. This dissertation concentrates on the study of self-calibrating parallel-link mechanisms. A framework of self-calibration of a parallel-link mechanism is created, which is based on kinematic analysis and the construction of measurement residuals utilizing the information provided by redundant sensors embedded in the system. Forward and inverse kinematic measurement residuals of the mechanisms are proposed. To avoid the estimation of redundant kinematic parameters of the mechanism, the concept of relative residuals is introduced. Guidelines for placement of sensors for self-calibration are presented. An approach to determining the number of independent kinematic parameters of the mechanism is introduced. Extensive simulation and experimental studies conducted on a parallel-link mechanism, the Stewart platform built in the Robotics Center at Florida Atlantic University, confirm the effectiveness of the proposed approach., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1997.

Summary: Vision systems have been widely used for parts inspection in electronics assembly lines. In order to improve the overall performance of a visual inspection system, it is important to employ an efficient object recognition algorithm. In this thesis work, a genetic algorithm based correlation algorithm is designed for the task of visual electronic parts inspection. The proposed procedure is composed of two stages. In the first stage, a genetic algorithm is devised to find a sufficient number of candidate image windows. For each candidate window, the correlation is performed between the sampled template and the image pattern inside the window. In the second stage, local searches are conducted in the neighborhood of these candidate windows. Among all the searched locations, the one that has a highest correlation value with the given template is selected as the best matched location. To apply the genetic algorithm technique, a number of important issues, such as selection of a fitness function, design of a coding scheme, and tuning of genetic parameters are addressed in the thesis. Experimental studies have confirmed that the proposed GA-based correlation method is much more effective in terms of accuracy and speed in locating the desired object, compared with the existing Monte-Carlo random search method., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.)--Florida Atlantic University, 1995.

Summary: Research and applications of genetic algorithms have become increasingly important in a wide variety of scientific fields. In this thesis, we present an empirical analysis of genetic algorithms in the function optimization area. As a focus of our research, a novel empirical analysis approach to various genetic algorithms is provided. The research starts from the survey of current trends in genetic algorithms, followed by exploring the characteristics of the simple genetic algorithm, the modified genetic algorithm and hybridized genetic algorithm. A number of typical function optimization problems are solved by these genetic algorithms. Ample empirical data associated with various modifications to the simple genetic algorithm is also provided. Results from this research can be used to assist practitioners in their applications of genetic algorithms., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.E.)--Florida Atlantic University, 1995.

Summary: The dual issues of extracting and tracking eye features from video images are addressed in this dissertation. The proposed scheme is different from conventional intrusive eye movement measuring system and can be implemented using an inexpensive personal computer. The desirable features of such a measurement system are low cost, accuracy, automated operation, and non-intrusiveness. An overall scheme is presented for which a new algorithm is forwarded for each of the function blocks in the processing system. A new corner detection algorithm is presented in which the problem of detecting corners is solved by minimizing a cost function. Each cost factor captures a desirable characteristic of the corner using both the gray level information and the geometrical structure of a corner. This approach additionally provides corner orientations and angles along with corner locations. The advantage of the new approach over the existing corner detectors is that it is able to improve the reliability of detection and localization by imposing criteria related to both the gray level data and the corner structure. The extraction of eye features is performed by using an improved method of deformable templates which are geometrically arranged to resemble the expected shape of the eye. The overall energy function is redefined to simplify the minimization process. The weights for the energy terms are selected based on the normalized value of the energy term. Thus the weighting schedule of the modified method does not demand any expert knowledge for the user. Rather than using a sequential procedure, all parameters of the template are changed simultaneously during the minimization process. This reduces not only the processing time but also the probability of the template being trapped in local minima. An efficient algorithm for real-time eye feature tracking from a sequence of eye images is developed in the dissertation. Based on a geometrical model which describes the characteris, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1994.

Summary: The focus of this thesis is the kinematic calibration of a SCARA arm with a hand-mounted camera. Kinematic calibration can greatly improve the accuracy of SCARA arms, which are widely used in electronic assembly lines. Vision-based robot calibration has the potential of being a fast, nonintrusive, low-cost, and autonomous approach. In this thesis, we apply a vision-based technique to calibrate SCARA arms. The robot under investigation is modeled by the modified complete and parametrically continuous model. By repeatedly calibrating the camera, the pose of the robot end-effector are collected at various robot measurement configurations. A least squares technique is then applied to estimate the geometric error parameters of the SCARA arm using the measured robot poses. In order to improve the robustness of the method, a new approach is proposed to calibrate the hand-mounted camera. The calibration algorithm is designed to deal with the case in which the camera sensor plane is nearly-parallel to the camera calibration board. Practical issues regarding robot calibration in general and SCARA arm calibration in particular are also addressed. Experiment studies reveal that the proposed camera-aided approach is a viable means for accuracy enhancement of SCARA arms., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.E.)--Florida Atlantic University, 1994.

Summary: Robot calibration using a vision system and moving cameras is the focus of this dissertation. The dissertation contributes in the areas of robot modeling, kinematic identification and calibration measurement. The effects of perspective distortion of circular camera calibration points is analyzed. A new modified complete and parametrically continuous robot kinematic model, an evolution of the complete and parametrically continuous (CPC) model, is proposed. It is shown that the model's error-model can be developed easily as the structure of this new model is very simple and similar to the Denavit-Hartenbert model. The derivation procedure of the error-model follows a systematic method that can be applied to any kind of robot arms. Pose measurement is the most crucial step in robot calibration. The use of stereo as well as mono mobile camera measurement system for collection of pose data of the robot end-effector is investigated. The Simulated Annealing technique is applied to the problem of optimal measurement configuration selection. Joint travel limits can be included in the cost function. It is shown that trapping into local minimum points can be effectively avoided by properly choosing an initial point and a temperature schedule. The concept of simultaneous calibration of camera and robot is developed and implemented as an automated process that determines the system model parameters using only the system's internal sensors. This process uses a unified mathematical model for the entire robot/camera system. The results of the kinematic identification, optimal configuration selection, and simultaneous calibration of robot and camera using the PUMA 560 robot arm have demonstrated that the modified complete and parametrically continuous model is a viable and simple modeling tool, which can achieve desired accuracy. The systematic way of modeling and performing of different kinds of vision-based robot applications demonstrated in this dissertation will pave th, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1993.

Summary: The objective of this dissertation is to develop effective algorithms for estimating the 3-D structure of a scene and its relative motion with respect to a camera or a pair of cameras from a sequence of images acquired by the cameras, under the assumption that the relative motion of the camera is small from one frame to another. This dissertation presents an approach of computing depth maps from an image sequence, which combines the direct depth estimation method with the optical flow based method. More specifically, optical flow on and near moving edges are computed using a correlation technique. The optical flow information is then fused with the gradient information to estimate depth not only on moving edges but also in internal regions. Depth estimation is formulated as a discrete Kalman filter problem and is solved in three stages. In the prediction stage, the depth map estimated for the previous frame, together with knowledge of the camera motion, is used to predict the depth variance at each pixel in the current frame. In the estimation stage, a vector-version of Kalman filter formulation is adapted and simplified to refine the predicted depth map. The resulting estimation algorithm takes into account the information from the neighboring pixels, and thus is much more robust than the scalar-version Kalman filter implementation. In the smoothing stage, morphological filtering is applied to reduce the effect of measurement noise and fill in uncertain areas based on the error covariance information. Since the depth at each pixel is estimated locally, the algorithm presented in this paper can be implemented on a parallel computer. The performance of the presented method is assessed through simulation and experimental studies. A new approach for motion estimation from stereo image sequences is also proposed in this dissertation. First a stereo motion estimation model is derived using the direct dynamic motion estimation technique. The problem is then solve, College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (Ph.D.)--Florida Atlantic University, 1992.

Summary: In order to facilitate the implementation of RAC-based camera calibration technique, several issues are addressed in this thesis. First, a straightforward extension of the RAC-based camera calibration technique to the case of unknown camera specifications is given. Second, to speed up the calibration process and reduce numerical difficulties, a simplified RAC-based method is presented. The simplified RAC-based method provides a closed-form solution to the camera calibration problem. Third, the PTM-based camera calibration technique is presented to give an example of pre-1985 camera calibration technique. Fourth, a method is devised to compute the ratio of scale factors. Finally, an optimization scheme is suggested to estimate image center. These modifications preserve all the advantages possessed by the original RAC-based calibration technique. The experiment results are provided to illustrate the effectiveness of the present methods., College of Engineering and Computer Science, Collection: FAU Electronic Theses and Dissertations Collection, Thesis (M.S.E.)--Florida Atlantic University, 1991.