Color Feature Extraction of Fingernail Image based on HSV Color Space as Early Detection Risk of Diabetes Mellitus

Abstract

Fingernail image color could be used for health diagnosis, such as detecting pancreatic condition as an indicator presence of diabetes mellitus risk. This paper focused on fingernail images as an early detection risk of diabetes mellitus. Therefore. the study aimed to analyze color features of fingernail images based on HSV (Hue, Saturation, Value) color space. The research data used fingernail images which were divided into three categories including normal, prediabetes, and diabetes data that according to blood glucose level. The data was cropped and extracted to each component of HSV color space. Analysis data was applied by grouping frequency distribution. The results revealed that among components of HSV, hue and value were overlapped between prediabetes and diabetes data. Component saturation had different range numbers in normal, prediabetes, and diabetes data. Therefore, it could be concluded that the HSV channel was considered as early detection of Diabetes Mellitus risk with fingernails image color as an object assay.