Name
Abstract | ||
|---|---|---|
Auricular acupressure therapy is a simple and effective means of health care and has gradually gained popularity. As an important prerequisite for auricular acupressure therapy, auricular point localization is difficult for general public. Most auricular points are distributed within eight regions, including helix, scapha, fossa triangularis, antihelix, concha, antitragus, tragus, and earlobes, which are divided according to the auricle's anatomical structure. In this paper, an improved YOLACT method is applied to segment the eight regions of the auricle automatically, and the accuracy is up to 93.2% on the ear dataset of 1000 images. Achieving segmentation of the auricular region, which can greatly narrow the localization area of auricular point, will provide an important foundation for the automatic localization of auricular point, favoring nonexperts to increase their understanding of auricular acupressure therapy or the development of an intelligent instrument related to auricular acupressure therapy. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1007/978-3-030-86608-2_8 | BIOMETRIC RECOGNITION (CCBR 2021) |
Keywords | DocType | Volume |
Auricular point localization, Ear region segmentation, Improved YOLACT | Conference | 12878 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hao Li | 1 | 25 | 11.35 |
| Xiaoyu Wang | 2 | 1487 | 69.46 |
| Zhichun Mu | 3 | 83 | 10.91 |