Gary L. Lebby | North Carolina Agricultural and Technical State University (original) (raw)
Papers by Gary L. Lebby
This paper presents a face detection system based on support vector machines (SVM). The lambda-co... more This paper presents a face detection system based on support vector machines (SVM). The lambda-coefficients (that correspond to the SVM support vectors) are determined from a set of training images from the Olivetti Research Lab (ORL) database. The support vectors are the archetypes for face and non-face images. Test images (containing both face and non-face images) are presented to the system. The SVM algorithm (now programmed and trained according to the support vector archetypes) maps the test images into higher dimension transform space where a hyperplane decision function is constructed. The hyperplane decision function is based on a degree-one polynomial (i.e. kernel function). The decision function is constructed equidistance between support vector archetypes to give an optimal hyperplane decision function. There are various methods for calculating the lambda-coefficients. In this paper, the coefficients are calculated using the multiplicative update algorithm proposed in (F. Sha, et al., Multiplicative Updates for Nonnegative Quadratic Programming in Support Vector Machines). Following the calculation of these coefficients and implementation of the SVM algorithm, we use the set of test images from the ORL database to demonstrate the SVM's performance as a face detection system. The goal or objective at this juncture is to simply determine the SVM's detection rate for the simplest kernel function (i.e. whether an image presented to the system is a face or non-face image)
Proceedings of the 6th International Symposium on Neural Networks Advances in Neural Networks Part Iii, 2009
Page 1. W. Yu, H. He, and N. Zhang (Eds.): ISNN 2009, Part III, LNCS 5553, pp. 11121121, 2009. ©... more Page 1. W. Yu, H. He, and N. Zhang (Eds.): ISNN 2009, Part III, LNCS 5553, pp. 11121121, 2009. © Springer-Verlag Berlin Heidelberg 2009 A Direct Approach to Achieving Maximum Power Conversion in Wind Power Generation Systems ...
Lecture Notes in Computer Science, 2010
Power System Load models have a wide range of application in the electric power industry includin... more Power System Load models have a wide range of application in the electric power industry including applications involving: (i) load management policy monitoring; (ii) assisting with the generator commitment problem; (iii) providing short term forecasts; (iv) aiding with system planning by providing long term forecasts. A method that has been utilized in the power systems planning community involves modeling the power system load (PSL) utilizing a truncated Fourier series. Presented herein is an innovative method based upon analyzing nine weeks of data and generating an optimum number of Fourier series terms included in model structure from a set of preselected heuristic basis functions for prediction. The resulting PSL model capable of providing high quality middle-long term forecasts and retain the shape prediction of the load curve out in time.
IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339), 1999
Page 1. Hopfield Like Networks for Pattern Recognition with Application to Face Recognition Dr. K... more Page 1. Hopfield Like Networks for Pattern Recognition with Application to Face Recognition Dr. Karl Ricanek, Jr. Dr. Gary L. Lebby Kirk Haywood Ricanek@ncat . edu Lebbv@ncat.edu Hawood@ncat . edu Department of Electrical ...
The 24th Southeastern Symposium on and The 3rd Annual Symposium on Communications, Signal Processing Expert Systems, and ASIC VLSI Design System Theory, 1992
The 24th Southeastern Symposium on and The 3rd Annual Symposium on Communications, Signal Processing Expert Systems, and ASIC VLSI Design System Theory, 1992
Proceedings Joint 9th IFSA World Congress and 20th NAFIPS International Conference (Cat. No. 01TH8569), 2001
Page 1. Self-organizing Map (SOM) Model for Mental Workload Classification Natalia Mazaeva, BA, C... more Page 1. Self-organizing Map (SOM) Model for Mental Workload Classification Natalia Mazaeva, BA, Celestine Ntuen, PbD. ... Scientific Promamming. Special Issue on Applications Analvsis, 4, 17 1 - 183. [ 121 Hecht-Nielsen, R. (1 987). Counterpropagation networks. ...
2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing, 2011
ABSTRACT
2008 40th Southeastern Symposium on System Theory (SSST), 2008
Proceedings of Thirtieth Southeastern Symposium on System Theory, 1998
Voltage collapse is the loss of stability in a power system due to a rapid decrease in voltage ca... more Voltage collapse is the loss of stability in a power system due to a rapid decrease in voltage causing blackouts. This paper discusses the prediction of voltage collapse of power systems. Algorithms are established which can be used to calculate the critical load point and provide information on system voltage variation with the load. A general transmission case is considered and a numerical example is presented
Proceedings of Thirtieth Southeastern Symposium on System Theory, 1998
1993 (25th) Southeastern Symposium on System Theory, 1993
Page 1. Design of an Expert System for Transmission Planning Gary L. Lebby and Lisa H. Antoine ..... more Page 1. Design of an Expert System for Transmission Planning Gary L. Lebby and Lisa H. Antoine ... 0 Graphical User Interface. Database Management System. Rule-Based Kernel. 0 Analytic Operational Evaluator. Capital Investment Evaluator. 0 Database Management System. ...
1993 (25th) Southeastern Symposium on System Theory, 1993
ABSTRACT
Smart Nondestructive Evaluation for Health Monitoring of Structural and Biological Systems, 2002
ABSTRACT Recent structural health monitoring techniques have focused on developing global sensor ... more ABSTRACT Recent structural health monitoring techniques have focused on developing global sensor systems that can detect damage on large structures. The approach presented here uses a piezoelectric sensor array system that mimics the biological nervous system architecture to measure acoustic emissions and dynamic strains in structures. The advantage of this approach is that the number of channels of data acquisition used for an N-by-N sensor array can be reduced from N2 to 2N. For large arrays the number of data acquisition channels is tremendously reduced. When transient damage events occur on the structure, the array output time histories can be recorded and the location of the excitation can be accurately determined using combinatorial logic. A trade-off is the difficulty of extracting individual sensor time histories from the array outputs without a neural network or a regressive technique. Only the sums of the sensor strains of each row and column can be exactly calculated using the voltage outputs of the array. The array approach allows efficient use of data acquisition instrumentation for structural health monitoring. Applications for the sensor array include crack and delamination detection, dynamic strain measurement, impact detection, and localization of damage on large complex structures.
2006 Proceeding of the Thrity-Eighth Southeastern Symposium on System Theory, 2006
... The decision function is constructed equidistance between support vector archetypes to give a... more ... The decision function is constructed equidistance between support vector archetypes to give an optimal ... algorithm proposed in [3]. Following the calculation of these coefficients andimplementation of the ... An overview of the various approaches to face detection can be found in [ ...
Journal of Bionic Engineering, 2010
Intelligent Automation & Soft Computing, 1997
ABSTRACT
Information Sciences, 1997
This paper presents a face detection system based on support vector machines (SVM). The lambda-co... more This paper presents a face detection system based on support vector machines (SVM). The lambda-coefficients (that correspond to the SVM support vectors) are determined from a set of training images from the Olivetti Research Lab (ORL) database. The support vectors are the archetypes for face and non-face images. Test images (containing both face and non-face images) are presented to the system. The SVM algorithm (now programmed and trained according to the support vector archetypes) maps the test images into higher dimension transform space where a hyperplane decision function is constructed. The hyperplane decision function is based on a degree-one polynomial (i.e. kernel function). The decision function is constructed equidistance between support vector archetypes to give an optimal hyperplane decision function. There are various methods for calculating the lambda-coefficients. In this paper, the coefficients are calculated using the multiplicative update algorithm proposed in (F. Sha, et al., Multiplicative Updates for Nonnegative Quadratic Programming in Support Vector Machines). Following the calculation of these coefficients and implementation of the SVM algorithm, we use the set of test images from the ORL database to demonstrate the SVM's performance as a face detection system. The goal or objective at this juncture is to simply determine the SVM's detection rate for the simplest kernel function (i.e. whether an image presented to the system is a face or non-face image)
Proceedings of the 6th International Symposium on Neural Networks Advances in Neural Networks Part Iii, 2009
Page 1. W. Yu, H. He, and N. Zhang (Eds.): ISNN 2009, Part III, LNCS 5553, pp. 11121121, 2009. ©... more Page 1. W. Yu, H. He, and N. Zhang (Eds.): ISNN 2009, Part III, LNCS 5553, pp. 11121121, 2009. © Springer-Verlag Berlin Heidelberg 2009 A Direct Approach to Achieving Maximum Power Conversion in Wind Power Generation Systems ...
Lecture Notes in Computer Science, 2010
Power System Load models have a wide range of application in the electric power industry includin... more Power System Load models have a wide range of application in the electric power industry including applications involving: (i) load management policy monitoring; (ii) assisting with the generator commitment problem; (iii) providing short term forecasts; (iv) aiding with system planning by providing long term forecasts. A method that has been utilized in the power systems planning community involves modeling the power system load (PSL) utilizing a truncated Fourier series. Presented herein is an innovative method based upon analyzing nine weeks of data and generating an optimum number of Fourier series terms included in model structure from a set of preselected heuristic basis functions for prediction. The resulting PSL model capable of providing high quality middle-long term forecasts and retain the shape prediction of the load curve out in time.
IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339), 1999
Page 1. Hopfield Like Networks for Pattern Recognition with Application to Face Recognition Dr. K... more Page 1. Hopfield Like Networks for Pattern Recognition with Application to Face Recognition Dr. Karl Ricanek, Jr. Dr. Gary L. Lebby Kirk Haywood Ricanek@ncat . edu Lebbv@ncat.edu Hawood@ncat . edu Department of Electrical ...
The 24th Southeastern Symposium on and The 3rd Annual Symposium on Communications, Signal Processing Expert Systems, and ASIC VLSI Design System Theory, 1992
The 24th Southeastern Symposium on and The 3rd Annual Symposium on Communications, Signal Processing Expert Systems, and ASIC VLSI Design System Theory, 1992
Proceedings Joint 9th IFSA World Congress and 20th NAFIPS International Conference (Cat. No. 01TH8569), 2001
Page 1. Self-organizing Map (SOM) Model for Mental Workload Classification Natalia Mazaeva, BA, C... more Page 1. Self-organizing Map (SOM) Model for Mental Workload Classification Natalia Mazaeva, BA, Celestine Ntuen, PbD. ... Scientific Promamming. Special Issue on Applications Analvsis, 4, 17 1 - 183. [ 121 Hecht-Nielsen, R. (1 987). Counterpropagation networks. ...
2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing, 2011
ABSTRACT
2008 40th Southeastern Symposium on System Theory (SSST), 2008
Proceedings of Thirtieth Southeastern Symposium on System Theory, 1998
Voltage collapse is the loss of stability in a power system due to a rapid decrease in voltage ca... more Voltage collapse is the loss of stability in a power system due to a rapid decrease in voltage causing blackouts. This paper discusses the prediction of voltage collapse of power systems. Algorithms are established which can be used to calculate the critical load point and provide information on system voltage variation with the load. A general transmission case is considered and a numerical example is presented
Proceedings of Thirtieth Southeastern Symposium on System Theory, 1998
1993 (25th) Southeastern Symposium on System Theory, 1993
Page 1. Design of an Expert System for Transmission Planning Gary L. Lebby and Lisa H. Antoine ..... more Page 1. Design of an Expert System for Transmission Planning Gary L. Lebby and Lisa H. Antoine ... 0 Graphical User Interface. Database Management System. Rule-Based Kernel. 0 Analytic Operational Evaluator. Capital Investment Evaluator. 0 Database Management System. ...
1993 (25th) Southeastern Symposium on System Theory, 1993
ABSTRACT
Smart Nondestructive Evaluation for Health Monitoring of Structural and Biological Systems, 2002
ABSTRACT Recent structural health monitoring techniques have focused on developing global sensor ... more ABSTRACT Recent structural health monitoring techniques have focused on developing global sensor systems that can detect damage on large structures. The approach presented here uses a piezoelectric sensor array system that mimics the biological nervous system architecture to measure acoustic emissions and dynamic strains in structures. The advantage of this approach is that the number of channels of data acquisition used for an N-by-N sensor array can be reduced from N2 to 2N. For large arrays the number of data acquisition channels is tremendously reduced. When transient damage events occur on the structure, the array output time histories can be recorded and the location of the excitation can be accurately determined using combinatorial logic. A trade-off is the difficulty of extracting individual sensor time histories from the array outputs without a neural network or a regressive technique. Only the sums of the sensor strains of each row and column can be exactly calculated using the voltage outputs of the array. The array approach allows efficient use of data acquisition instrumentation for structural health monitoring. Applications for the sensor array include crack and delamination detection, dynamic strain measurement, impact detection, and localization of damage on large complex structures.
2006 Proceeding of the Thrity-Eighth Southeastern Symposium on System Theory, 2006
... The decision function is constructed equidistance between support vector archetypes to give a... more ... The decision function is constructed equidistance between support vector archetypes to give an optimal ... algorithm proposed in [3]. Following the calculation of these coefficients andimplementation of the ... An overview of the various approaches to face detection can be found in [ ...
Journal of Bionic Engineering, 2010
Intelligent Automation & Soft Computing, 1997
ABSTRACT
Information Sciences, 1997