Feature-based marker detection for body height estimation through key point extraction and image rectification
Body height is one of the most common biometrics measured in an individual. Height measurement, when done manually and repeatedly for a large population with limited manpower, the task can be time consuming. This situation is not unusual in the context of telehealth, especially in rural areas. Thus,...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | text |
| Published: |
Animo Repository
2023
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/faculty_research/13059 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Summary: | Body height is one of the most common biometrics measured in an individual. Height measurement, when done manually and repeatedly for a large population with limited manpower, the task can be time consuming. This situation is not unusual in the context of telehealth, especially in rural areas. Thus, there arises a need for an automated height estimation system. This research tackles the development of an accessible height estimation system that only requires a single marker of known height. The system utilizes computer vision and image processing techniques to extract necessary key points to estimate the height of a person from an image. The system provided two height estimates: one directly estimated from the height of the image, and another wherein an attempt to correct the projection distortion through image rectification was performed. The experiments have found that the estimations with rectification tend to be more accurate on average but have higher variance (0.05cm ± 11.04) as opposed to without rectification (3.34cm ± 5.85). |
|---|