Discipline: Computer Science
The purpose of this study was to optimize the use of the Eigenface together with the Baye’s Theorem in recognizing the facial expression by using another method, which is the Simplex Method. Facial expression recognition is a research topic with interesting applications in the field of human-computer interaction and psychology. The classification accuracy for this system, which uses static images as input has a large limitation by the image quality, the lighting conditions and the orientation of the depicted face and deformation of the facial features. These problems are partially overcome by using a combined method of Eigenface and Bayes’ Theorem. Because the rule Bayes’ Theorem is to only process the closest probability of the result or the likelihood combined by the prior condition over the marginal likelihood, it is still limited in its function as mentioned in the latter part. While the first two methods are used in the facial expression recognition, another method is used in optimizing the application by doing the function on determining the most number of facial expressions in the static image with one up to five persons, and that method is the Simplex Method. The system classifies expressions from one of the emotional categories happy, anger, sadness, surprise, and fear with remarkable accuracy and limitations.