Semester: 10th Semester, Master Thesis Title
Download 0.97 Mb.
|
- Navigate this page:
- 3.5.1.Emotions
- 3.5.2.Emotional Intelligence
- 3.5.3.Computer Vision
- 3.5.4.Affective Computing
- 4.Test method
- 4.1.Test Strategy
- 4.1.1.Test Phases
3.5.Analysis ConclusionThe following will briefly outline the specific knowledge gained from the different topics from the analysis chapter. This will lead to a requirement specification for the overall completion of this project based on the recommendations and conclusions from the selected analysed articles. The following is divided into four separate categories, each representing topics that will guide the test phase, design phase and implementation phase of this project: 3.5.1.EmotionsEmotions are of a highly subjective nature, both from the interpreter and the perceiver’s point of view. Ekman found that emotions within the semantic primitives were generally universally understood; different cultures viewed and understood emotions the same way. Ekman and Keltner found that only emotions elicited by the human face abode to the rule, emotions conveyed orally or via body language did not. Keating investigated which facial features improved the attractiveness of an individual and found that common denominators existed across cultures. Certain facial features enhanced the attractiveness whereas others diminished it. Therefore this thesis believes that emotions to a certain extent are universally understood in regards to the basic primitives. Therefore by only focusing on the human smile, which studies found signalled happiness, an average understanding of what the smile constitutes can be found with the assistance of a test group. It will be the task of the test group to determine how much a person is smiling. 3.5.2.Emotional IntelligenceThe level of emotional intelligence in an individual greatly influences their perception and interaction with peers. In the studies by Kirsten Boehner she generalised emotions, specifically the smile as being universally understood as a sign of happiness. In the research of Emotional Intelligence studies has shown, that the level of EI in females where, on average, higher than their male counterparts. Therefore as a precaution the test phase of this thesis will contain a separation of gender to be able to investigate, if the eventual should occur, differences between male and female ratings. Lastly if the ratings between genders do differentiate, it will show in a higher rating of smile in the test results in the group with the suspected higher level of emotional intelligence. To note is that an individual with a high level of emotional intelligence is faster and more accurate in determining the emotional state of the individual he or she is observing. Furthermore the location in which the test will be conducted is important due to the fact, that the environment can have an effect on the emotional perception. Therefore this thesis believes that the level of emotional intelligence in the individuals conducting the test can influence the test results. Test subjects will be divided into two groups, male and female, should the test results reveal a significant difference in their ratings. Lastly the test will take place in a location selected by the test participants; therefore a website containing the test will be created to suit this requirement. Furthermore test of Emotional Intelligence showed that the display of emotional states via pictures assisted the test subjects the most in determining the emotional state of the individual, test subjects responded most fondly to emotional display via pictures. 3.5.3.Computer VisionThe recommendations in the computer vision chapter focused on the picture databases used for both testing the algorithm as well as the test subjects. To avoid pictures with imposed emotional displays and so forth, a picture set will be found only containing randomly selected pictures. The requirement for the pictures will only be that the person in the picture will display either happiness or sadness or a gradient between the two categorisations. 3.5.4.Affective ComputingKirsten Boehner and Maja Pantic recommended that the location in which the test will take place be carefully considered. The location can have both a positive and negative effect on the emotional state of the individual taking the test. This further substantiates the decision to create a website for the test thus allowing the test subjects to choose their own test location, hopefully eliminating an environment that would influence their test answers. 4.Test methodThe following chapter will describe and outline the process the accompanying test to this thesis underwent. The chapter will include the recommendations from the analysis chapter on how to obtain the most concise results and how to conduct the specifics of the test. It will furthermore describe the choices made in deciding how the test was delivered to the test subjects and which concerns were alleviated by the chosen test method. 4.1.Test StrategyThe overall test strategy was derived from the questions proposed in the final problem statement, namely how a human understands a smile. Before a computer can be tasked with rating the level of smile a person is conveying, human trials will have to be conducted. The human test subjects will be tasked to rate the level of smile in pictures presented to them. The results obtained from the human trials will be used to guide the computer in an attempt to reach the same accuracy and understanding as the test subjects. 4.1.1.Test Phases
Directory: projekter -> files files -> IT’d be weird without mcdonald’S files -> Of Maj 2014 aau introduction & Theory files -> Representations of Gender in Agatha Christie’s Poirot and Miss Marple files -> Research in contemporary social movements: a case study of Guatemala 2015 files -> Anne Hviid-Pilgaard Master Thesis 31/05 2012 Table of Contents files -> Jakob Nors-Ganer Aalborg University 28-05-2015 files -> Abstract: Purpose files -> Security and trust in IoT/M2m cloud based platform files -> The impact of french and british colonial rule on the he system in cameroon from the perspective of students files -> Rumble in the jungle the ‘Blessing’ and ‘Curse’ of Mineral Wealth in the Congo Download 0.97 Mb. Share with your friends:
|