A fuzzy-logic-based model for assessment of human facial harmony based on theory of averages

• Published in: Journal of the Franklin Institute Vol. 348; no. 8; pp. 2216 - 2234
• Main Authors: Ozkul, Murat Haluk, Ozkul, Tarik
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01-10-2011; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Assessments; Biological and medical sciences; Blends; Computer aided testing; Computer science; control theory; systems; Control theory. Systems; Exact sciences and technology; Facial; Human; Mathematical models; Medical sciences; Modelling and identification; Patients; Pattern recognition. Digital image processing. Computational geometry; Surgeons; Surgery (general aspects). Transplantations, organ and tissue grafts. Graft diseases; Human; Fuzzy logic; Software tool; Face recognition; Surgery; Facies; Observer; Aesthetics; Logical programming; System identification; Modeling; Logic model
• ISSN: 0016-0032; 1879-2693
• DOI: 10.1016/j.jfranklin.2011.06.014

What makes some human faces more attractive than others is an interesting phenomenon which is difficult to explain based on few facial features. Research indicated that many facial parameters contribute to generation of a “harmonious face” which is found appealing by most observers. There are different theories developed on which and how these parameters blend in to make a harmonious face. Determination of these parameters and the underlying mechanism is important since achieving a harmonious face is the primary goal in dental and aesthetic surgical procedures. This article presents a model developed for computer-aided assessment of facial harmony, which assesses the facial harmony of male patients and generates a score that reflect the level of harmony. The computer-aided tool presented in this paper uses “theory of averages” as a mechanism for measurement of facial harmony which is frequently used by aesthetic surgeons and orthodontists for assessment of facial proportions of the patient. The output of the developed model is an index which reflects the degree of harmony of facial features and how well they blend into each other.