Related Experiment Video
Updated: Aug 14, 2026
05:49
An Automated Squint Method for Time-syncing Behavior and Brain Dynamics in Mouse Pain Studies
Published on: November 1, 2024
Optimized High- Resolution Network for Accurate Facial Wrinkle Detection Using Harbor Seal Whiskers Optimization
V Senthil Murugan1, V Hemasree2, Wulfran Fendzi Mbasso3,4
1Department of Computer Science and Engineering, St. Joseph's College of Engineering, Chennai, India.
Biomedical Engineering and Computational Biology
|August 13, 2026
Summary
This study introduces a novel facial wrinkle detection framework (FWD-DHRN-HSWOA) that enhances image quality and uses high-resolution features for accurate wrinkle localization. The method shows promising results for automated facial aging assessment and skincare applications.
Area of Science:
- Computer Vision
- Medical Imaging
- Artificial Intelligence
Background:
- Automated facial wrinkle detection is crucial for applications like facial-ageing assessment, cosmetic analysis, and dermatological screening.
- Wrinkles present challenges due to their low contrast, irregular structure, and susceptibility to variations in lighting, pose, skin texture, and image quality.
Purpose of the Study:
- To propose a novel facial wrinkle-detection framework (FWD-DHRN-HSWOA) integrating advanced techniques for improved accuracy.
- To combine attentional image enhancement, high-resolution feature representation, and metaheuristic parameter optimization for robust wrinkle detection.
Main Methods:
- Facial images from the FG-NET Aging Database were utilized and augmented with custom wrinkle annotations.
- Image preprocessing involved alignment, resizing, normalization, and enhancement using Deep Attentional Guided Image Filtering.
- A Dynamic Lightweight High-Resolution Network was employed for fine spatial information preservation, with parameters optimized via Harbor Seal Whiskers Optimization.
Main Results:
- The proposed FWD-DHRN-HSWOA framework demonstrated superior performance compared to baseline models in terms of accuracy, precision, recall, F1-score, and specificity.
- The framework achieved a lower Root Mean Square Error (RMSE) in wrinkle detection within the tested FG-NET-derived annotation setting.
- Results highlight performance under specific experimental conditions and caution against generalization to diverse demographic groups or datasets.
Conclusions:
- The integration of attentional preprocessing, high-resolution feature extraction, and metaheuristic optimization shows significant potential for precise facial wrinkle localization.
- Further external validation on diverse, well-characterized wrinkle datasets is essential for the broad clinical and commercial deployment of the framework.