Quantification of Pigmented Regions in Detracted Images

Abstract

This study presents a method for analyzing and quantifying pigmented areas on dental surfaces after the application of caries detection products. The methodology integrates image enhancement, the creation of binary masks, and conversion to the HSV color space to isolate and accurately calculate the extent of pigmentation. A dataset consisting of 200 images of unpigmented teeth and 200 images of pigmented teeth was used, with strict criteria for selecting images that provided a clear frontal view, focusing on the upper and lower canine regions. The image processing steps included using LabelMe software for manual annotation and applying binary masks to segment the teeth from the background. Specific color filters identified pigmented regions. Histogram analysis in the HSV space validated the predominance of red shades corresponding to the pigmented areas, confirming the accuracy of the segmentation. The results demonstrate the effectiveness of the proposed method for calculating pigmented areas, providing an objective measure that can be used to assess the efficacy of products designed for bacterial plaque detection in dental applications.