Smile challenger

• Main Author: Ang, Shi Chao
• Other Authors: Ravi Suppiah
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision; DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: http://hdl.handle.net/10356/70212
• Institution: Nanyang Technological University

A smile is a form of facial expression that are normally closely associated with happiness. Given that a smile can be measured, can we measure happiness and turn it into a fun Smile Challenger game? This paper presents an efficient approach to analyze a smile and assign a reasonable score (0 – 100%) to it. The study adopted a Mouth-Corner Features (MCFs)-based algorithm in determining the intensity of a smile. Through digital image processing techniques, it made use of a deterministic Inverse Binary Thresholding method to extract the MCFs from the image. With the two corners of the mouth detected, the algorithm then computes the Smile Score. The Smile Score, which categorized different classification of a smile, is consisted of 85% Corner Width Score and 15% Wrinkle Density Score. Smile Challenger’s algorithm was tested with 300 face images and had achieved an accuracy 71.7%.