(Institut Teknologi Sepuluh Nopember, Indonesia)(2) * Nanik Suciati
(Institut Teknologi Sepuluh Nopember, Indonesia)*corresponding author
AbstractFacial wrinkle segmentation is critical for automated dermatological assessment, yet existing deep learning methods exhibit significant performance degradation under real-world illumination and pose variations, restricting practical clinical deployment where imaging conditions cannot be controlled. This study proposes a novel robustness-oriented segmentation framework that integrates three synergistic components: (1) attention-enhanced U-Net architectures with strategically frozen VGG16/VGG19 backbones enabling hierarchical feature transfer, (2) a dual augmentation strategy coupling geometric transformations for pose invariance with a four-level photometric enhancement pipeline for illumination robustness, and (3) a weighted mask fusion mechanism combining expert annotations with weak supervision labels. Three architectures (baseline Attention U-Net, VGG16, and VGG19 variants) are trained on 1,000 FFHQ-Wrinkle images and systematically evaluated across four augmentation strategies under nine challenging deployment conditions, including low light, high contrast, noise, head tilts, and perspective shifts. The proposed VGG19 Attention U-Net with combined augmentation achieves a Dice coefficient of 0.6533 and IoU of 0.4931, outperforming the best existing method (Striped WriNet) by +4.26% in Dice and +5.89% in IoU under identical re-implemented training conditions. The model retains 97.82% of its original performance across all nine perturbation conditions (robustness score: 0.6391), representing a 10.4% robustness improvement over the non-augmented baseline. These results demonstrate that the synergistic combination of attention mechanisms, transfer learning, and dual augmentation produces clinically viable robustness for facial wrinkle segmentation.
KeywordsAttention U-Net;Data Augmentation;Facial Wrinkle Segmentation;Robustness Evaluation;VGG Backbone
|
DOIhttps://doi.org/10.26555/ijain.v12i3.2394 |
Article metricsAbstract views : 171 | PDF views : 2 |
Cite |
Full Text Download
|
References
[1] M. H. Yap, N. Batool, C.-C. Ng, M. Rogers, and K. Walker, “A Survey on Facial Wrinkles Detection and Inpainting: Datasets, Methods, and Challenges,” IEEE Trans. Emerg. Top. Comput. Intell., vol. 5, no. 4, pp. 505–519, Aug. 2021, doi: 10.1109/TETCI.2021.3075723.
[2] R. Soekarta and K. R. Ku-Mahamud, “Recent Facial Image Preprocessing Techniques: A Review,” in The 8th Mechanical Engineering, Science and Technology International Conference, Basel Switzerland: MDPI, Feb. 2025, p. 39. doi: 10.3390/engproc2025084039.
[3] E. McMullen et al., “Machine learning methods for determining skin age: A systematic review,” J. Tissue Viability, vol. 34, no. 3, p. 100887, Aug. 2025, doi: 10.1016/j.jtv.2025.100887.
[4] G. Carlos da Silva, M. B. Barbosa, F. B. C. Júnior, P. L. Moreira, R. Werka, and A. A. Martin, “Detection of skin wrinkles and quantification of roughness using a novel image processing technique from a dermatoscope device,” Ski. Res. Technol., vol. 29, no. 6, p. e13335, Jun. 2023, doi: 10.1111/srt.13335.
[5] R. Pooth, S. Sattler, F. Westerberg, T. Pavicic, and M. Kerscher, “Transforming Skin Quality Evaluation With: From Subjective Grading to Data‐Driven Precision,” J. Cosmet. Dermatol., vol. 24, no. S4, p. e70371, Sep. 2025, doi: 10.1111/jocd.70371.
[6] I. Ahmad and F. Alqurashi, “Early cancer detection using deep learning and medical imaging: A survey,” Crit. Rev. Oncol. Hematol., vol. 204, no. December, p. 104528, Dec. 2024, doi: 10.1016/j.critrevonc.2024.104528.
[7] H. Yoon, S. Kim, J. Lee, and S. Yoo, “Deep-Learning-Based Morphological Feature Segmentation for Facial Skin Image Analysis,” Diagnostics, vol. 13, no. 11, p. 1894, May 2023, doi: 10.3390/diagnostics13111894.
[8] S. S. Kumar, R. S. Vinod Kumar, and D. Subbulekshmi, “A Review of U‐Net‐Based Deep Learning Models for Skin Lesion Segmentation,” Int. J. Imaging Syst. Technol., vol. 35, no. 3, p. e70107, May 2025, doi: 10.1002/ima.70107.
[9] H. M. Balaha, E. R. Antar, M. M. Saafan, and E. M. El-Gendy, “A comprehensive framework towards segmenting and classifying breast cancer patients using deep learning and Aquila optimizer,” J. Ambient Intell. Humaniz. Comput., vol. 14, no. 6, pp. 7897–7917, Jun. 2023, doi: 10.1007/s12652-023-04600-1.
[10] W. F. Setiawan, A. Amirullah, I. P. Ariatama, and R. N. E. Anggraini, “CNN-LSTM Architecture for Multi-Task Sentiment and Emotion Classification on Large-Scale Indonesian TikTok Application Reviews,” CommIT (Communication Inf. Technol. J., vol. 20, no. 1, p. 77, Mar. 2026, doi: 10.21512/commit.v20i1.13876.
[11] E. Jelli et al., “Single‐cell segmentation in bacterial biofilms with an optimized deep learning method enables tracking of cell lineages and measurements of growth rates,” Mol. Microbiol., vol. 119, no. 6, pp. 659–676, Jun. 2023, doi: 10.1111/mmi.15064.
[12] A. Kirillov et al., “Segment Anything,” in 2023 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, Oct. 2023, pp. 3992–4003. doi: 10.1109/ICCV51070.2023.00371.
[13] J. Chen, M. He, and W. Cai, “Facial Wrinkle Detection with Multiscale Spatial Feature Fusion Based on Image Enhancement and ASFF-SEUnet,” Electronics, vol. 12, no. 24, p. 4897, Dec. 2023, doi: 10.3390/electronics12244897.
[14] S. Kim, H. Yoon, J. Lee, and S. Yoo, “Facial wrinkle segmentation using weighted deep supervision and semi-automatic labeling,” Artif. Intell. Med., vol. 145, no. November, p. 102679, Nov. 2023, doi: 10.1016/j.artmed.2023.102679.
[15] M.-Y. Yang, Q.-L. Shen, D.-T. Xu, X.-L. Sun, and Q.-B. Wu, “Striped WriNet: Automatic wrinkle segmentation based on striped attention module,” Biomed. Signal Process. Control, vol. 90, no. April, p. 105817, Apr. 2024, doi: 10.1016/j.bspc.2023.105817.
[16] H. Kim, J. Lee, J. Rew, and E. Hwang, “Automated Facial Wrinkle Segmentation Scheme Using UNet++,” KSII Trans. Internet Inf. Syst., vol. 18, no. 8, pp. 2333–2345, Aug. 2024, doi: 10.3837/tiis.2024.08.015.
[17] J. Moon, H. Chung, and I. Jang, “Facial Wrinkle Segmentation for Cosmetic Dermatology: Pretraining with Texture Map-Based Weak Supervision,” in Pattern Recognition, A. Antonacopoulos, S. Chaudhuri, R. Chellappa, C.-L. Liu, S. Bhattacharya, and U. Pal, Eds., Cham: Springer Nature Switzerland, 2025, pp. 319–334. doi: 10.1007/978-3-031-78104-9_22.
[18] C. Turner et al., “Deep learning predicted perceived age is a reliable approach for analysis of facial ageing: A proof of principle study,” J. Eur. Acad. Dermatology Venereol., vol. 38, no. 12, pp. 2295–2302, Dec. 2024, doi: 10.1111/jdv.20365.
[19] J. Cheng, H. Li, D. Li, S. Hua, and V. S. Sheng, “A Survey on Image Semantic Segmentation Using Deep Learning Techniques,” Comput. Mater. Contin., vol. 74, no. 1, pp. 1941–1957, 2023, doi: 10.32604/cmc.2023.032757.
[20] O. Oktay et al., “Attention U-Net: Learning Where to Look for the Pancreas,” arXiv Prepr., vol. 3, pp. 1–10, 2018, [Online]. Available: https://arxiv.org/abs/1804.03999
[21] W. Zeng, “Image data augmentation techniques based on deep learning: A survey,” Math. Biosci. Eng., vol. 21, no. 6, pp. 6190–6224, 2024, doi: 10.3934/mbe.2024272.
[22] Z. Liu, Q. Qi, S. Wang, and G. Zhai, “A novel approach to the detection of facial wrinkles: Database, detection algorithm, and evaluation metrics,” Comput. Biol. Med., vol. 174, no. May, p. 108431, May 2024, doi: 10.1016/j.compbiomed.2024.108431.
[23] A. T. Tran, T. Zeevi, and S. Payabvash, “Strategies to Improve the Robustness and Generalizability of Deep Learning Segmentation and Classification in Neuroimaging,” BioMedInformatics, vol. 5, no. 2, p. 20, Apr. 2025, doi: 10.3390/biomedinformatics5020020.
[24] E. Rahimi, C. Li, X. Zhong, G. H. Shi, and A. M. Ardekani, “The role of initial lymphatics in the absorption of monoclonal antibodies after subcutaneous injection,” Comput. Biol. Med., vol. 183, no. December, p. 109193, Dec. 2024, doi: 10.1016/j.compbiomed.2024.109193.
[25] S. Gkikas and M. Tsiknakis, “Automatic assessment of pain based on deep learning methods: A systematic review,” Comput. Methods Programs Biomed., vol. 231, no. April, p. 107365, Apr. 2023, doi: 10.1016/j.cmpb.2023.107365.
[26] R. M. Elbashir and M. Hoon Yap, “Evaluation of Automatic Facial Wrinkle Detection Algorithms,” J. Imaging, vol. 6, no. 4, p. 17, Apr. 2020, doi: 10.3390/jimaging6040017.
[27] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” in Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015, N. Navab, J. Hornegger, W. M. Wells, and A. F. Frangi, Eds., Cham: Springer International Publishing, 2015, pp. 234–241. doi: 10.1007/978-3-319-24574-4_28.
[28] A. Li, D. Li, and A. Wang, “A Two-Stage YOLOv5s–U-Net Framework for Defect Localization and Segmentation in Overhead Transmission Lines,” Sensors, vol. 25, no. 9, p. 2903, May 2025, doi: 10.3390/s25092903.
[29] R. Jiao et al., “Learning with limited annotations: A survey on deep semi-supervised learning for medical image segmentation,” Comput. Biol. Med., vol. 169, no. February, p. 107840, Feb. 2024, doi: 10.1016/j.compbiomed.2023.107840.
[30] A. Asperti and D. Filippini, “Deep Learning for Head Pose Estimation: A Survey,” SN Comput. Sci., vol. 4, no. 4, p. 349, Apr. 2023, doi: 10.1007/s42979-023-01796-z.
[31] Y. Cai, H. Bian, J. Lin, H. Wang, R. Timofte, and Y. Zhang, “Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement,” in 2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023, pp. 12470–12479. doi: 10.1109/ICCV51070.2023.01149.
[32] R. Azad et al., “Medical Image Segmentation Review: The Success of U-Net,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 46, no. 12, pp. 10076–10095, Dec. 2024, doi: 10.1109/TPAMI.2024.3435571.
[33] S. ATASEVER, N. AZGINOGLU, D. S. TERZI, and R. TERZI, “A comprehensive survey of deep learning research on medical image analysis with focus on transfer learning,” Clin. Imaging, vol. 94, pp. 18–41, Feb. 2023, doi: 10.1016/j.clinimag.2022.11.003.
[34] A. Ghaznavi, R. Rychtáriková, P. Císař, M. M. Ziaei, and D. Štys, “Symmetry Breaking in the U-Net: Hybrid Deep-Learning Multi-Class Segmentation of HeLa Cells in Reflected Light Microscopy Images,” Symmetry (Basel)., vol. 16, no. 2, p. 227, Feb. 2024, doi: 10.3390/sym16020227.
[35] K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., vol. 6, pp. 1–14, Apr. 2015, [Online]. Available: http://arxiv.org/abs/1409.1556
[36] J. Deng, W. Dong, R. Socher, L.-J. Li, Kai Li, and Li Fei-Fei, “ImageNet: A large-scale hierarchical image database,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Jun. 2009, pp. 248–255. doi: 10.1109/CVPR.2009.5206848.
[37] P. Svobodova, K. Sethia, P. Strakos, and A. Varysova, “Automatic Hepatic Vessels Segmentation Using RORPO Vessel Enhancement Filter and 3D V-Net with Variant Dice Loss Function,” Appl. Sci., vol. 13, no. 1, p. 548, Dec. 2022, doi: 10.3390/app13010548.
[38] R. Azad et al., “Loss Functions in the Era of Semantic Segmentation: A Survey and Outlook,” vol. 1, pp. 1–24, Dec. 2023, [Online]. Available: http://arxiv.org/abs/2312.05391
[39] S. Jadon, “A survey of loss functions for semantic segmentation,” in 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), IEEE, Oct. 2020, pp. 1–7. doi: 10.1109/CIBCB48159.2020.9277638.
[40] L. Wang et al., “Improving adversarial robustness of deep neural networks by using semantic information,” Knowledge-Based Syst., vol. 226, p. 107141, Aug. 2021, doi: 10.1016/j.knosys.2021.107141.
[41] M. Dhiaf, A. C. Rouhou, Y. Kessentini, and S. Ben Salem, “MSdocTr-Lite: A lite transformer for full page multi-script handwriting recognition,” Pattern Recognit. Lett., vol. 169, pp. 28–34, May 2023, doi: 10.1016/j.patrec.2023.03.020.
[42] G. Bana, F. Lamadie, S. Charton, T. Randriamanantena, D. Lucor, and N. Sheibat-Othman, “BYG-drop: a tool for enhanced droplet detection in liquid–liquid systems through machine learning and synthetic imaging,” Front. Chem. Eng., vol. 6, pp. 1–14, Aug. 2024, doi: 10.3389/fceng.2024.1415453.
[43] C. Budd, L. C. Garcia-Peraza Herrera, M. Huber, S. Ourselin, and T. Vercauteren, “Rapid and robust endoscopic content area estimation: a lean GPU-based pipeline and curated benchmark dataset,” Comput. Methods Biomech. Biomed. Eng. Imaging Vis., vol. 11, no. 4, pp. 1215–1224, Jul. 2023, doi: 10.1080/21681163.2022.2156393.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0