A Comparison between visual shade matching and digital shade analysis system using K-NN algorithm

Abstract

Objective: One of the major challenges in dentistry is the correct selection of tooth color as close to the natural adjacent tooth color as possible so this study was directed to develop an alternative way to traditional methods for shade matching. Subjects and Methods: A smartphone application was developed, a total of 1300 shade tab images (50 per shade tab) were captured using a smartphone camera with auto-mode settings and special ring light which emits 5500 k light temperature. The images were shot at angled distances of 14 – 20 cm from a shade guide. Color features were extracted and classified using K nearest neighbor classification algorithm (K-NN) to form a software training dataset. Sixty two upper central incisors were captured by smartphone and the shade of these teeth were matched using developed software and, visual methods compere to the spectrophotometer. Results: The matching accuracy using software gives 70% matching compare to the visual method which gives 65 % matching accuracy . Conclusions : Shade matching using a Smartphone application and Digital photography can emerge as a viable alternative to the use of spectrophotometers for shade selection in a clinical setup.