Improved high-resolution network for palmprint principal line extraction

Wei Jia; Xin Zhou

doi:10.1117/12.3032972

5 July 2024 Improved high-resolution network for palmprint principal line extraction

Wei Jia, Xin Zhou

Proceedings Volume 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024); 131845L (2024) https://doi.org/10.1117/12.3032972
Event: 3rd International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 2024, Kuala Lumpur, Malaysia

Abstract

Recently, there has been growing interest in deep learning-based semantic segmentation. We propose a high-resolution network UC-HRNet for palmprint principal line extraction. Compared with other popular deep learning semantic segmentation networks, the performance is stronger. Dense prediction tasks rely heavily on both high-resolution imagery and sophisticated semantic representations for optimal performance. Generally, low-resolution feature maps excel in conveying robust semantic representations, whereas high-resolution feature maps excel in capturing fine-grained details like edges but may lack in semantic richness. Current cutting-edge frameworks, like HRNet, preserves the parallelism between low- and high-resolution feature maps and iteratively exchanges information across different resolutions. The success of UNet is mainly due to its U-shaped structure as well as jump connections.The symmetric U-shaped structure allows for a more thorough fusion of front and back features, it turns out that skip connections are effective in recovering complex details of target objects. We propose a kind of U-shaped structure based on HRNet and add skip connections to make it able to segment the target more accurately. We utilize deep supervision to acquire hierarchical representations from fully aggregated feature maps. Compared to other popular semantic segmentation networks, UC-HRNet proves its effectiveness on three master libraries of semantically segmented palmprints created by ourselves.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Wei Jia and Xin Zhou "Improved high-resolution network for palmprint principal line extraction", Proc. SPIE 13184, Third International Conference on Electronic Information Engineering and Data Processing (EIEDP 2024), 131845L (5 July 2024); https://doi.org/10.1117/12.3032972

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
9 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Semantics

Databases

Image segmentation

Printing

Image filtering

Feature extraction

Feature fusion

Show All Keywords

Keywords/Phrases

Search In:

Publication Years