<![CDATA[The skin is one of the essential and vital organs or parts of the human body and is a mirror of health and life. However, sometimes people still know about facial skin problems and visit a dermatologist. Examinations are usually carried out conventionally by doctors to analyze facial skin problems. However, doctors need time to diagnose facial skin problems, which can cause long queues. Since the COVID-19 outbreak, all activities have been restricted. Positive cases of COVID -19 which continue to increase every day in Indonesia make direct interaction between doctors and patients very dangerous. In this study, a classification of types of facial skin problems will be carried out using digital image processing technology. Image analysis using the Histogram of Oriented Gradient (HOG) method, feature extraction is obtained by calculating the gradient of the image to be grouped and block normalization will be carried out. The HOG parameters used are pixels per cell = (8, 8), orientation = 9, cells per block = (3, 3), and block norm = L2-Hys. Then the classification process uses the Support Vector Machine method to identify facial skin problems, namely black spots, acne and wrinkles. Classification of facial skin problems using the SVM method gets an accuracy value of 98% with a processing time of 8.79 seconds.]]>
Copyrights © 2021
