Classification of Facial Acne Types Based on Self-Supervised Learning using DINOv2
DOI:
https://doi.org/10.30871/jaic.v10i1.11856Keywords:
DINOv2, Self-Supervised Learning, LoRA, Acne Type Classification, Computer VisionAbstract
Acne is a common inflammatory skin condition that can affect an individual’s psychological well-being and overall quality of life. The inability to independently recognize specific types of acne often leads to the use of inappropriate skincare products. This situation highlights the need for an image-based classification system that can provide accurate visual identification. The self-supervised learning method Distillation with NO Labels, version 2 (DINOv2), is employed as a feature extractor to classify four types of acne—Acne fulminans, Acne nodules, Papules, and Pustules—using the “skin-90” dataset. The fine-tuning process is conducted through a Parameter-Efficient Fine-Tuning (PEFT) approach using Low-Rank Adaptation (LoRA) to adjust the model’s visual representations to the acne domain without updating all parameters in full, followed by integration with a classification head. The results show that the model achieves an accuracy of 90.70%, with precision, recall, and F1-score values of 90.64%, 90.68%, and 90.57%, respectively. The findings suggest that the proposed architectural design and training configuration are suitable for capturing relevant visual patterns of acne, while further validation is required to assess robustness across more diverse data distributions.
Downloads
References
[1] G. V. Agustin, M. Ayub, and S. L. Liliawati, “Deteksi dan Klasifikasi Tingkat Keparahan Jerawat: Perbandingan Metode You Only Look Once,” J. Tek. Inform. dan Sist. Inf., vol. 10, no. 3, pp. 2443–2229, 2024, [Online]. Available: http://dx.doi.org/10.28932/jutisi.v10i3.9414
[2] A. Quattrini, C. Boër, T. Leidi, and R. Paydar, “A Deep Learning-Based Facial Acne Classification System,” Clin. Cosmet. Investig. Dermatol., vol. 15, no. April, pp. 851–857, 2022, doi: 10.2147/CCID.S360450.
[3] A. Beauty and Y. Erlyana, “Perancangan Buku Berjudul ‘My Acne Journey, Kenali, Atasi, Cegah Jerawat’ Untuk Remaja Usia 13-18 Tahun Dengan Teknik Ilustrasi Digital,” J. Dimens. DKV Seni Rupa dan Desain, vol. 7, no. 2, pp. 143–162, 2022, doi: 10.25105/jdd.v7i2.12813.
[4] I. G. Ayu, Y. Wahyuning, N. Made, S. Noviana, and N. T. Aryanata, “Hubungan Antara Citra Tubuh Dengan Kepercayaan Diri Pada Model Di Agency X makan dikarenakan keinginan besar untuk mengakibatkan terjadinya gangguan citra Citra Tubuh dengan Kepercayaan Diri pada Model di Agency X ”. Penelitian ini bertujuan untuk mengeta,” vol. 4, no. 1, pp. 8–13, 2023.
[5] E. M. Sipayung and E. Christopher R., “Klasifikasi Image Jenis Kayu pada Furnitur dengan Convolutional Neural Network,” J. Telemat., vol. 18, no. 2, pp. 82–87, 2024, doi: 10.61769/telematika.v18i2.617.
[6] T. Matius and S. Mulyana, “Penentuan Area Wajah Menggunakan Algoritma Viola-Jones Untuk Aplikasi Manipulasi Warna Citra [ Face Area Determination Using Viola-Jones Algorithm For Digital Image Colour Manipulation Application ],” J. Algoritm. Log. dan Komputasi, no. 01, pp. 743–750, 2025.
[7] L. Jing and Y. Tian, “Self-Supervised Visual Feature Learning with Deep Neural Networks: A Survey,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 43, no. 11, pp. 4037–4058, 2021, doi: 10.1109/TPAMI.2020.2992393.
[8] T. Uelwer et al., “A Survey on Self-Supervised Representation Learning,” pp. 1–48, 2023, [Online]. Available: http://arxiv.org/abs/2308.11455
[9] M. Baharoon, W. Qureshi, J. Ouyang, Y. Xu, A. Aljouie, and W. Peng, “Towards General Purpose Vision Foundation Models for Medical Image Analysis: An Experimental Study of DINOv2 on Radiology Benchmarks,” pp. 1–22, 2023, [Online]. Available: http://arxiv.org/abs/2312.02366
[10] T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, “A simple framework for contrastive learning of visual representations,” 37th Int. Conf. Mach. Learn. ICML 2020, vol. PartF16814, no. Figure 1, pp. 1575–1585, 2020.
[11] K. He and F. Ai, “Momentum Contrast for Unsupervised Visual Representation Learning”.
[12] J. Grill et al., “Bootstrap Your Own Latent A New Approach to Self-Supervised Learning,” vol. 200.
[13] J. Mohan, A. Sivasubramanian, S. V., and V. Ravi, “Enhancing skin disease classification leveraging transformer-based deep learning architectures and explainable AI,” Comput. Biol. Med., vol. 190, pp. 1–17, 2025, doi: 10.1016/j.compbiomed.2025.110007.
[14] A. R. Rafif Raif, R. E. Putra, A. Prapanca, and A. Qoiriah, “Penerapan DINOv2 pada Content Based Image Retrieval (CBIR) dalam Website Katalog Digital Batik Surabaya,” J. Informatics Comput. Sci., vol. 6, no. 03, pp. 678–687, 2024, doi: 10.26740/jinacs.v6n03.p678-687.
[15] B. Cui, M. Islam, L. Bai, and H. Ren, “Surgical-DINO: adapter learning of foundation models for depth estimation in endoscopic surgery,” Int. J. Comput. Assist. Radiol. Surg., vol. 19, no. 6, pp. 1013–1020, 2024, doi: 10.1007/s11548-024-03083-5.
[16] I. G. N. L. W. K. Virna Dalira Br Sebayang, “Klasifikasi Jenis Jerawat Berdasarkan Citra Menggunakan Convolutional Neural Network dengan Arsitektur MobileNetV2,” J. FASILKOM, vol. 14, no. 2, pp. 766–774, 2024, doi: 10.56211/helloworld.v3i2.518.
[17] A. Christopher and T. M. S. Mulyana, “Klasifikasi Tumbuhan Angiospermae Menggunakan Algoritma K-Nearest Neighbor Berdasarkan Pada Bentuk Daun,” JIPI (Jurnal Ilm. Penelit. dan Pembelajaran Inform., vol. 7, no. 4, pp. 1233–1243, 2022, doi: 10.29100/jipi.v7i4.3211.
[18] C. N. Putri, W. D. Qornain, F. Bamahri, G. E. Yuliastuti, and M. Kurniawan, “Klasifikasi Jenis Jerawat pada Data Citra Jerawat Wajah Menggunakan Convolutional Neural Network,” TIN Terap. Inform. Nusant., vol. 5, no. 2, pp. 172–181, 2024, doi: 10.47065/tin.v5i2.5231.
[19] J. Woźna, K. Korecka, J. Stępka, A. Bałoniak, R. Żaba, and R. A. Schwartz, “Acne fulminans treatment: case report and literature review,” Front. Med., vol. 11, no. July, pp. 1–7, 2024, doi: 10.3389/fmed.2024.1450666.
[20] Y. Li, X. Hu, G. Dong, X. Wang, and T. Liu, “Acne treatment: research progress and new perspectives,” Front. Med., vol. 11, no. July, pp. 1–9, 2024, doi: 10.3389/fmed.2024.1425675.
[21] A. Paichitrojjana, “Malassezia Folliculitis: A Review Article,” J. Med. Assoc. Thail., vol. 105, no. 2, pp. 160–167, 2022, doi: 10.35755/jmedassocthai.2022.02.13268.
[22] D. This, “Understanding OOD detection methods for image classification networks Eindhoven University of Technology Data Science and Artificial Intelligence,” 2024.
[23] S. D. Pratama, L. Lasimin, and M. N. Dadaprawira, “Pengujian Black Box Testing Pada Aplikasi Edu Digital Berbasis Website Menggunakan Metode Equivalence Dan Boundary Value,” J-SISKO TECH (Jurnal Teknol. Sist. Inf. dan Sist. Komput. TGD), vol. 6, no. 2, p. 560, 2023, doi: 10.53513/jsk.v6i2.8166.
[24] R. V Reynolds et al., “FROM THE ACADEMY Executive summary: Guidelines of care for the management of acne vulgaris,” J Am Acad Dermatol, vol. 90, pp. 1006–1010, 2024, [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/38300170/
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Gantari Chardaputeri, Angelina Pramana Thenata
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
