Skin Issue Detection System Using Machine Learning and Computer Vision

ABSTRACT The increasing use of cosmetic products has raised concerns about their potential adverse effects on facial skin health. This system aims to develop an automated system to detect facial skin problems caused by cosmetic products. The system will utilize facial images to identify common skin issues such as acne, redness, rashes, pigmentation, and other dermatological conditions. Data pre-processing techniques will be used to enhance image quality and reduce noise. Feature extraction methods will be used to isolate critical skin attributes, followed by the application of supervised learning algorithms for classification. The system will be trained on a labelled dataset of images, where each image is associated with a specific type of skin problem and a history of cosmetic product usage. Additionally, the project will include exploratory data analysis to understand correlations between different cosmetic ingredients and skin issues. The model's performance will be evaluated using metrics such as accuracy, precision, recall, and F1-score to ensure high detection capabilities. The ultimate goal is to provide users tool that can help identify potential skin problems early, and the causes related to cosmetic products. To make this system user-friendly and accessible, it will be developed as a mobile or web application. The application will allow users to upload images of their faces, which will then be analyzed in real-time. Users will receive instant feedback on potential skin issues and recommendations based on detected conditions. By transforming it into a mobile or web application, the system becomes more accessible, enabling individuals to monitor their skin health conveniently. This innovation has the potential to revolutionize skincare by helping users make informed choices about the cosmetic products they use. Keywords: adverse effects, automated system, cosmetic ingredients, data pre-processing, dermatological conditions, feature extraction.