Recognition of human face expressions

Download
index.pdf
2006
Ener, Emrah
Show full item record
172
views
56
downloads
In this study a fully automatic and scale invariant feature extractor which does not require manual initialization or special equipment is proposed. Face location and size is extracted using skin segmentation and ellipse fitting. Extracted face region is scaled to a predefined size, later upper and lower facial templates are used for feature extraction. Template localization and template parameter calculations are carried out using Principal Component Analysis. Changes in facial feature coordinates between analyzed image and neutral expression image are used for expression classification. Performances of different classifiers are evaluated. Performance of proposed feature extractor is also tested on sample video sequences. Facial features are extracted in the first frame and KLT tracker is used for tracking the extracted features. Lost features are detected using face geometry rules and they are relocated using feature extractor. As an alternative to feature based technique an available holistic method which analyses face without partitioning is implemented. Face images are filtered using Gabor filters tuned to different scales and orientations. Filtered images are combined to form Gabor jets. Dimensionality of Gabor jets is decreased using Principal Component Analysis. Performances of different classifiers on low dimensional Gabor jets are compared. Feature based and holistic classifier performances are compared using JAFFE and AF facial expression databases.
Applications of Electric Power.
http://etd.lib.metu.edu.tr/upload/3/12607521/index.pdf
https://hdl.handle.net/11511/16289
Graduate School of Natural and Applied Sciences, Thesis

Suggestions

Application of a natural-resonance based feature extraction technique to small-scale aircraft wires for electromagnetic target classification
Ersoy, Mehmet Okan; Sayan, Gönül; Department of Electrical and Electronics Engineering (2004)
The problem studied in this thesis, is the classification of the small-scale aircraft targets by using a natural resonance based electromagnetic feature extraction technique. The aircraft targets are modeled by perfectly conducting, thin wire structures. The electromagnetic back-scattered data used in the classification process, are numerically generated for five aircraft models. A contemporary signal processing tool, the Wigner-Ville distribution is employed in this study in addition to using the principal...
Series active filter design, control, and implementation with a novel load voltage harmonic extraction method
Şentürk, Osman Selçuk; Hava, Ahmet Masum; Department of Electrical and Electronics Engineering (2007)
Series Active Filters (SAF) are designed for harmonic isolation and load voltage regulation of single-phase and three-phase voltage harmonic source type nonlinear loads. The novel Absolute Value Method (AVM) for load voltage harmonic extraction is proposed and applied in the control algorithm of SAF. The SAF compensated systems are represented by simplified linear models such that SAF controller gains can be easily determined. Harmonic isolation and load voltage regulation performances of 2.5 kW single-phas...
Optimizing transient and filtering performance of a C-type 2nd harmonic power filter by the use of solid-state switches
Gerçek, Cem Özgür; Ertaş, Arif; Department of Electrical and Electronics Engineering (2007)
In this research work, the performance of a C-type, 2nd harmonic power filter is optimized by the use of a thyristor switched damping resistor. In the design of conventional C-type, 2nd harmonic filters; the resistance of permanently connected damping resistor is to be optimized for minimization of voltage stresses on filter elements arising from switchings in transient state and for maximization of filtering effectiveness in the steady-state. Transformer inrush current during energization of power transfor...
Design and implementation of advanced pulse width modulation techniques and passive filters for voltage source inverter driven three-phase ac motors
Nebi, Onur Çetin; Hava, Ahmet Masum; Department of Electrical and Electronics Engineering (2010)
Advanced pulse width modulation (PWM) techniques such as space vector PWM, active zero state PWM, discontinuous PWM, and near state PWM methods are used in three-phase AC motor drives for the purpose of obtaining low PWM current ripple, wide voltage linearity range, and reduced common mode voltage (CMV). In some applications, a filter is inserted between the inverter and the motor for the purpose of reducing the stresses in the motor. The motor current PWM ripple components, terminal voltage overshoots, sha...
Repetitive control of a three-phase uninterruptible power supply with isolation transformer
Çetinkaya, Süleyman; Hava, Ahmet Masum; Department of Electrical and Electronics Engineering (2006)
A repetitive control method for output voltage control of a three phase uninterruptible power supply (UPS) with isolation transformer is investigated. In the method voltage control loop is employed in the stationary dq frame. The controller eliminates the periodic errors on the output voltages due to inverter voltage nonlinearity and load disturbances. The controller design and implementation details are given. The controller is implemented on a 5-kVA UPS prototype which is constructed in laboratory. Linear...
Citation Formats
E. Ener, “Recognition of human face expressions,” M.S. - Master of Science, Middle East Technical University, 2006.
METU IIS - Integrated Information Service open@metu.edu.tr