Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
SPECIAL

OpenCV를 이용한 부위별 안면마비 등급 측정 시스템 설계

Design of Facial Paralysis Class Measurement System Using OpenCV

Journal of the Korean Society for Precision Engineering 2023;40(7):533-538.
Published online: July 1, 2023

1 강원대학교 대학원 스마트헬스과학기술융합학과

1 Department of Smart Health Science and Technology, Graduate School, Kangwon National University

#E-mail: yjpark@kangwon.ac.kr, TEL: +82-033-250-6371
• Received: April 28, 2023   • Accepted: June 8, 2023

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 7 Views
  • 0 Download
  • 1 Crossref
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • A Review on Development Trends of Facial Palsy Grading System: Mainly on Automatic Method
    Ja-Ha Lee, Jeong-Hyun Moon, Gyoungeun Park, Won-Suk Sung, Young-soo Kim, Eun-Jung Kim
    Korean Journal of Acupuncture.2025; 42(1): 1.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Design of Facial Paralysis Class Measurement System Using OpenCV
J. Korean Soc. Precis. Eng.. 2023;40(7):533-538.   Published online July 1, 2023
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Design of Facial Paralysis Class Measurement System Using OpenCV
J. Korean Soc. Precis. Eng.. 2023;40(7):533-538.   Published online July 1, 2023
Close

Figure

  • 0
  • 1
  • 2
  • 3
Design of Facial Paralysis Class Measurement System Using OpenCV
Image Image Image Image
Fig. 1 Output results for Nottingham System scale and House-Brackmann grade. (a) Inserted Pictures, (b) System output results
Fig. 2 Example of displaying face landmarks for virtual models using the OpenCV system. (a) Normal, (b) Mild facial asymmetry, (c) Severe facial asymmetry
Fig. 3 Measurement locations for the Nottingham System (NS)
Fig. 4 Additional measurement locations for facial muscles
Design of Facial Paralysis Class Measurement System Using OpenCV

Measurement results for the Nottingham System and HBGS grade

No. Side Raise eyebrow Close eye tightly Smile Sum Nottingham Scale [%] House-Brackmann grade
ΔSO to IO from rest [pixel] ΔSO to IO from rest [pixel] ΔLC to M from rest [pixel]
1 Left 25.3 20.7 38.6 84.6 97.1 2
Right 28 18.2 36 82.1
2 Left 29.8 6.3 15.9 52.1 99.4 2
Right 24.7 9.5 17.6 51.8
3 Left 19.9 22.6 19.6 62.1 97.9 2
Right 22 18 23.4 63.4
4 Left 27.6 16.5 16.5 60.6 68.7 3
Right 22 10 9.6 41.6
5 Left 1.5 12 23 37.1 64.8 3
Right 4.5 4.9 14.5 24
6 Left 1.4 5.9 15.7 23 84.8 2
Right 1.9 7.4 17.8 27.1
7 Left 20.5 1.5 26 48.5 54.4 3
Right 36.5 22 30.6 89
8 Left 0.6 15.6 0.7 17 26.4 4
Right 17.4 15.6 31.5 64.4
9 Left 5 5.6 2.8 13.4 24.9 5
Right 1.4 17 35.4 53.7

Improved eyebrow and mouth corner muscle measurement results for each of the three points, and grade of facial muscle movement

No. Side Raise eyebrow+Close eye tightly Smile Grade of facial muscle movement
ΔEyebrow to IO [pixel] ΔMouth to LC [pixel]
Point 1 Point 2 Point 3 Point 1 Point 2 Point 3 Eyebrow Mouth
1 Left 37.09 44.94 44.55 38.62 27.24 34.55 1 1
Right 38.54 47.41 46.31 35.96 26.36 31.80 1 1
2 Left 28.31 33.57 35.80 15.93 18.83 15.87 3 3
Right 31.80 36.14 35.10 14.86 17.58 14.54 3 4
3 Left 36.41 44.88 43.88 19.59 23.98 17.65 2 3
Right 27.46 37.01 38.78 23.42 26.82 20.75 2 2
4 Left 37.53 38.57 37.84 16.51 16.15 17.12 2 3
Right 31.88 36.26 38.15 9.60 6.97 9.67 2 4
5 Left 14.23 12.89 11.80 23.60 15.35 21.58 4 3
Right 10.90 12.54 12.50 14.55 7.41 11.54 4 4
6 Left 8.21 15.27 7.58 17.81 15.53 16.09 4 3
Right 7.88 12.70 7.43 15.70 10.83 9.78 4 4
7 Left 20.53 28.37 32.68 26.43 26.30 26.36 3 2
Right 40.38 51.15 48.44 30.58 27.87 27.55 1 2
8 Left 14.88 5.85 1.80 0.75 5.87 1.14 5 5
Right 41.48 46.64 43.40 31.45 28.21 30.06 1 1
9 Left 5.13 5.60 3.17 2.85 5.77 1.41 5 5
Right 27.16 31.59 28.92 35.38 33.38 35.37 3 1

Newly measured facial asymmetry scale

No. Eyebrow [%] Mouth [%] Total [%]
Point 1 Point 2 Point 3 Point 1 Point 2 Point 3
1 96.2 94.8 96.2 93.1 96.8 92.1 93.5
2 89.0 92.9 98.0 93.3 93.3 91.6 90.3
3 75.4 82.5 88.4 83.6 89.4 85.1 79.5
4 84.9 94.0 99.2 58.1 43.1 56.5 64
5 76.6 97.3 94.4 61.6 48.3 53.5 62.5
6 96.0 83.1 97.9 88.2 69.7 60.7 71.9
7 50.8 55.5 67.5 86.4 94.4 95.7 68.6
8 35.9 12.5 4.2 2.4 20.8 3.8 3.3
9 18.9 17.7 11.0 8.0 17.3 4.0 7.5

Comparison of measurement results between the previous system and our system

No. Our system The previous system
Grade of facial muscle movement Facial asymmetry scale [%] HBGS NS scale [%] HBGS
Left Right
1 1 1 93.5 2 97.1 2
2 3 4 90.3 2 99.4 2
3 3 2 79.5 2 97.9 2
4 3 3 64 3 68.7 3
5 4 4 62.5 3 64.8 3
6 4 4 71.9 3 84.8 2
7 3 2 68.6 3 54.4 3
8 5 1 3.3 5 26.4 4
9 5 2 7.5 5 24.9 5
Table 1 Measurement results for the Nottingham System and HBGS grade
Table 2 Improved eyebrow and mouth corner muscle measurement results for each of the three points, and grade of facial muscle movement
Table 3 Newly measured facial asymmetry scale
Table 4 Comparison of measurement results between the previous system and our system