A Web App for Acne Severity Identification with RGB Image Color Scheme

This study addresses the need to detect the severity of acne in skin care and beauty, particularly in managing the diagnosis of acne severity more efficiently and objectively. Acne, a common skin problem, can significantly affect an individual's quality of life and self-confidence. Referring to the prevalence of acne, this study leverages advancements in Deep Learning and Convolutional Neural Networks (CNN) to design a classification model capable of identifying the severity of acne in facial images using the RGB color scheme. CNN is an artificial neural network architecture that can process image data efficiently and accurately. CNN can extract essential features from acne images, such as color, texture, and shape, and classify the severity of acne into four categories: level 0, level 1, level 2, and level 3. The simple application of the model not only provides an efficient solution for acne diagnosis but also has the potential to improve consistency and objectivity in healthcare services. By incorporating transfer learning and color schemes (RGB), the testing results show that the model successfully classifies the severity of acne with an accuracy of 86.89%. Thus, this research contributes to technical and technological advancements and has the potential to positively impact the overall quality of facial skin care services, marking a significant first step in improving facial skin care services.