A challenge of deep‐learning‐based object detection for hair follicle dataset
Journal of Cosmetic Dermatology · 2023-09 · 22 (9) · 2565-2578
DOI: 10.1111/jocd.15742 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Cosmetic Dermatology, online erschienen am 06.04.2023. Verfasst von Wei Lv und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Alibaba Cloud Big Data Application, Zhuhai College of Science and Technology, Zhuhai, China. Hauptthemen laut MEDLINE: Hair Follicle, Deep Learning. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangdong Basic and Applied Basic Research Foundation, Guangdong Key Disciplines Project, Natural Science Foundation of Guangdong Province.
Abstract (Originalsprache)
Background
Deep‐learning object detection has been applied in various industries, including healthcare, to address hair loss.
Methods
In this paper, YOLOv5 object detection algorithm was used to detect hair follicles in a small and specific image dataset collected using a specialized camera on the scalp of individuals with different ages, regions, and genders. The performance of YOLOv5 was compared with other popular object detection models.
Results
YOLOv5 performed well in the detection of hair follicles, and the follicles were classified into five classes based on the number of hairs and the type of hair contained. In single‐class object detection experiments, a smaller batch size and the smallest YOLOv5s model achieved the best results, with an map of 0.8151. In multiclass object detection experiments, the larger YOLOv5l model was able to achieve the best results, and batch size affected the result of model training.
Conclusion
YOLOv5 is a promising algorithm for detecting hair follicles in a small and specific image dataset, and its performance is comparable to other popular object detection models. However, the challenges of small‐scale data and sample imbalance need to be addressed to improve the performance of target detection algorithms.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Alibaba Cloud Big Data Application, Zhuhai College of Science and Technology, Zhuhai, China.
ORCID-Profile: Wei Lv, Yifan Zeng, Weihong Liu, Chuying Huang
Förderung
- Guangdong Basic and Applied Basic Research Foundation
- Guangdong Key Disciplines Project
- Natural Science Foundation of Guangdong Province
- Zhuhai College of Science and Technology 2021 Innovation and Entrepreneurship Training Program for College Students
Zitieren
APA
Lv, W., Chen, T., Zeng, Y., Liu, W., & Huang, C. (2023). A challenge of deep‐learning‐based object detection for hair follicle dataset. Journal of Cosmetic Dermatology, 22(9), 2565-2578. https://doi.org/10.1111/jocd.15742
Vancouver
Lv W, Chen T, Zeng Y, Liu W, Huang C. A challenge of deep‐learning‐based object detection for hair follicle dataset. Journal of Cosmetic Dermatology. 2023;22(9):2565-2578. doi:10.1111/jocd.15742
BibTeX
@article{Lv2023,
author = {Lv, Wei and Chen, Tao and Zeng, Yifan and Liu, Weihong and Huang, Chuying},
title = {A challenge of deep‐learning‐based object detection for hair follicle dataset},
journal = {Journal of Cosmetic Dermatology},
year = {2023},
volume = {22},
number = {9},
pages = {2565--2578},
doi = {10.1111/jocd.15742}
}
EUVIMED