Abstract
dc:description.abstractComputerized emotion recognition systems can be powerful tools to help solve problems in a wide range of fields including education, healthcare, and marketing. Existing systems use digital images or live video to track facial expressions on a person's face and deduce that person's emotional state. The research presented in this thesis explores combinations of several facial feature extraction techniques with different classifier algorithms. Namely, the feature extraction techniques used in this research were Discrete Cosine/Sine Transforms, Fast Walsh-Hadamard Transform, Principle Component Analysis, and a novel method called XPoint. Features were extracted from both global (using the entire facial image) and local (using only facial regions like the mouth or eyes) contexts and classified with Linear Discriminant Analysis and k-Nearest Neighbor algorithms. Some experiments also fused many of these features into one system in an effort to create even more accurate systems. The system accuracy for each feature extraction method/classifier combination was calculated and discussed. The combinations that performed the best produced systems between 85%-90% accurate. The most accurate systems utilized Discrete Sine Transform from global and local features in a Linear Discriminant Analysis classifier, as well as feature fusion of all features in a Linear Discriminant Classifier.
Degree
thesis:*- Name thesis:degree_name
- M.S. Electrical Engineering
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Electrical and Computer Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Smirnov, Demiyan
- Contributors dc:contributor
-
- Ramachandran, Ravi
- Head, Linda
- Thayasivam, Uma
Subjects
dc:subject × 8Identifiers
dc:identifier.*- Repository record dc:identifier
- https://rdw.rowan.edu/etd/2705