Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{177678,
        author = {Priyanshu Singh and T Tanusree and Munzir Hassan and Karan Kumar Das},
        title = {Smart Dermatology: Enhancing Skin Type Identification and Product Recommendation using Computer Vision},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {6108-6115},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=177678},
        abstract = {This research, we introduce a machine
learning based personalized skincare recommendation
system which classifies facial skin into the three basic
types i.e.: Oily, Dry and Normal by using deep learning.
To improve generalization, the facial images were
processed with resizing, normalization and data
augmentation to create a curated dataset. The system
takes advantage of CNNs and pretrain model such as
ResNet50, EfficientNet-B0 using ensemble learning to
obtain high accuracy based on robust feature extraction.
With F1-score of 94% and classification accuracy of 92%,
the hybrid model combining ResNet50 and EfficientNetB0 bested by performance, but also by the fact that it can
capture dermatological nuances regardless of lighting or
skin tone. This model also outperforms typical ML models
like SVM, RF, and KNN in terms of accuracy and has
strong promise for deployment in real time mobile
skincare apps and virtual dermatology tools where the
current primary solutions are unacceptable},
        keywords = {Skin products, ResNet50, EfficientNet-B0, Artificial Intelligence, Dermatology},
        month = {May},
        }