Facial Skin Classification by using Machine Learning Approach

Read full text Summarise with VIRAT
Authors: Varadarajan Pranav, Akkineni Saras Chandrika, Gudikati Thanmai, Mandapalli Ruthvik, Ramesh Athe
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Background: The classification of facial skin types accurately is crucial for dermatologyand cosmetics, but traditional approaches depend on personal assessments that are capableof errors. An alternative which is more accurate and objective is machine-learning.Objective: This research uses deep-learning and feature extraction techniques to create anautomated machine learning-based method for classifying the various types of facial skin:normal/dry/oily.Methods: We used 166 facial images as our dataset, which were manually collected. Weused preprocessing-methods such as CLAHE for contrast enhancement and Gaussian-Blurand Bilateral-Filtering for noise reduction to prepare the data. MobileNetV2 and a HOGextracted features. We have used the Adam optimizer, categorical cross-entropy loss, andthe deep-learning models ResNet50 and VGG16 to classify the images. 5-fold crossvalidation was used to evaluate-performance.

Fields of Study & Subjects

Face recognition and analysis Advanced Neural Network Applications Artificial Intelligence in Healthcare

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
1 January 2026
Language
EN