Facial wrinkle is one of the most prominent biological changes thataccompanying the natural aging process. However, there are some externalfactors contributing to premature wrinkles development, such as sun exposureand smoking. Clinical studies have shown that heavy smoking causes prematurewrinkles development. However, there is no computerised system that canautomatically assess the facial wrinkles on the whole face. This studyinvestigates the effect of smoking on facial wrinkling using a social habitface dataset and an automated computerised computer vision algorithm. Thewrinkles pattern represented in the intensity of 0-255 was first extractedusing a modified Hybrid Hessian Filter. The face was divided into tenpredefined regions, where the wrinkles in each region was extracted. Then thestatistical analysis was performed to analyse which region is effected mainlyby smoking. The result showed that the density of wrinkles for smokers in tworegions around the mouth was significantly higher than the non-smokers, atp-value of 0.05. Other regions are inconclusive due to lack of large scaledataset. Finally, the wrinkle was visually compared between smoker andnon-smoker faces by generating a generic 3D face model.
展开▼