Dorsal hand vein segmentation using Vein-Generative Adversarial Network (V-GAN) Model
Difficulty in achieving intravenous access in some patients is a clinical problem due to extreme age, body size, and chronic disease patients. In biometric identification, hand vein patterns are useful when other external identifiers are more prone to be damaged or forged. To overcome these problems...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
Springer Science and Business Media Deutschland GmbH
2022
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/39631/1/Dorsal%20Hand%20Vein%20Segmentation%20Using%20Vein-Generative%20Adversarial.pdf http://umpir.ump.edu.my/id/eprint/39631/2/Dorsal%20hand%20vein%20segmentation%20using%20Vein-Generative%20Adversarial%20Network%20%28V-GAN%29%20Model_ABS.pdf http://umpir.ump.edu.my/id/eprint/39631/ https://doi.org/10.1007/978-981-16-8129-5_89 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Difficulty in achieving intravenous access in some patients is a clinical problem due to extreme age, body size, and chronic disease patients. In biometric identification, hand vein patterns are useful when other external identifiers are more prone to be damaged or forged. To overcome these problems, near-infrared dorsal hand vein images are captured and segmented for vein extraction. However, the segmentation process becomes more challenging when the infrared im- ages suffer from extremely low contrast and distortion, indirectly affecting the segmentation process. Therefore, this work presents a method for generating an accurate map of dorsal hand vein patterns using deep learning Vein-Generative Adversarial Networks (V-GAN). The performance of V-GAN is measured in terms of accuracy, Area under Curve (AUC), F1-score, sensitivity, specificity, and dice-coefficient. |
|---|