An Automatic Identification System of Human Skin Irritation

Quantitative characterization of human skin irritation is an important but difficult task. Recently, identification of human skin is still doing manually. Furthermore, manual identification of the human skin irritation sample can be very subjective. The skin irritation analysis could conducted using biochemical test, but not simple. In this research, a new approach an automatic human skin identification system base on image recognition have been developed. Skin image processed using pattern recognition methods to obtain decision of skin sample test is skin irritation or not. System design is implementation of Gray Level Histogram (GLH) feature or texture Gray Level Co-occurrence Matrices (GLCM) features using classifier distance metric: Manhattan distance and Euclidean distance, or Learning Vector Quantization (LVQ) neural network. Combination between feature extrator and classifier methods proposed to evaluate of performance system. The experimental results show that the best accuracy namely 83.33% obtained when design system is implementation of GLH or GLCM features using LVQ neural network classifier.