LUTBIO multimodal biometric database

The LUTBIO database provides a comprehensive resource for research in multimodal biometric authentication, featuring the following key aspects:

- Extensive Biometric Modalities: The database contains data from nine biometric modalities: voice, face, fingerprint, contact-based palmprint, electrocardiogram (ECG), opisthenar (back of hand), ear, contactless palmprint, and periocular region.

- Diverse Demographics: Data were collected from 306 individuals, with a balanced gender distribution of 164 males and 142 females, spanning an age range of 8 to 90 years. This diverse age representation enables analyses across a wide demographic spectrum.

- Representative Population Sampling: Volunteers were recruited from naturally occurring communities, ensuring a large-scale, statistically representative population. The collected data encompass variations observed in real-world environments.

- Support for Multimodal and Cross-Modality Research: LUTBIO provides both contact-based and contactless palmprint data, as well as fingerprint data (from optical images and scans), promoting advancements in multimodal biometric authentication. This resource is designed to guide the development of future multimodal databases.

- Flexible, Decouplable Data: The biometric data in the LUTBIO database are designed to be highly decouplable, enabling independent processing of each modality without loss of information. This flexibility supports both single-modality and multimodal analysis, empowering researchers to optimize, combine, and customize biometric features for specific applications.

✅ Data Availability: If you wish to use the LUTBIO dataset, please download the attached Word document, fill in the information, and send it as an attachment to rykeryang AT 163.com. We will process your request as soon as possible!

🥸 Important Notice: Please read the data collection protocol of the LUTBIO dataset carefully before use, as it is essential for understanding and correctly interpreting the dataset. Thank you.

😎 Good news! Our paper has been accepted by Information Fusion, and the DOI is https://doi.org/10.1016/j.inffus.2025.102945. We appreciate the reviewers and the editor for their efforts.🥰🥰🥰