Here we share the NTU datasets to all the researchers who are working on biometrics and forensics field.
The NTU datasets consist of the following sub-datasets:
NTUIIS contains two parts: NTUIIS_1 and NTUIIS_2. Images in NTUIIS_1 were collected for visual effect evaluation. To generate original (uncompressed) images for comparisons, they were resized to remove the compression artifacts and referred to as references. Images were further downscaled to a low resolution so that blood vessels were clear in the original images and JPEG compression artifacts were observable in the compressed images with the compression factors of 50 and 75. There are 44 images and 40 images in NTUIIS_1 for testing JPEG compression factors of 50 and 75, respectively. Images in NTUIIS_2 were collected for objective evaluation based on blood vessel matching. For each image in NTUIIS_2, there is one corresponding image from the same body part of the same subject. The forearms were extracted and formed two datasets: a left forearm dataset and a right forearm dataset. Their sizes are 156 and 162 images from 78 and 81 subjects respectively. Their quality factors range from 50 to 90.
Please cite the following paper to use the NTU IIS dataset:
Xiaojie Li and Adams Kong, “A Multi-model Restoration Algorithm for Recovering Blood Vessels in Skin Images”, Image and Vision Computing, vol. 61, pp. 40-53, 2017
Human back skin dataset contains 699 back skin images collected in Biometrics and Forensics Lab, SCSE, NTU.
Image number: 699 (52 random pose images and 647 standard pose images)
Image type: jpg
Please cite the following paper to use the dataset:
-
Nhat Quang Huynh, Xingpeng Xu, Adams Kong and Sathyan Subbiah, “A preliminary report on a full-body imaging system for effectively collecting and processing biometric traits of prisoners”, IEEE Symposium Series on Computational Intelligence, 2014.
-
Arfika Nurhudatiana and Adams Kong, "On Criminal Identification in Color Skin Images Using Skin Marks (RPPVSM) and Fusion with Vein Patterns", IEEE Transactions on Information Forensics and Security, vol. 10, no. 5, pp. 916-931, 2015
Human chest skin dataset contains 434 chest skin images collected in Biometrics and Forensics Lab, SCSE, NTU.
Image number: 434
Image type: jpg
Please cite the following paper to use the dataset:
-
Nhat Quang Huynh, Xingpeng Xu, Adams Kong and Sathyan Subbiah, “A preliminary report on a full-body imaging system for effectively collecting and processing biometric traits of prisoners”, IEEE Symposium Series on Computational Intelligence, 2014.
-
Arfika Nurhudatiana and Adams Kong, "On Criminal Identification in Color Skin Images Using Skin Marks (RPPVSM) and Fusion with Vein Patterns", IEEE Transactions on Information Forensics and Security, vol. 10, no. 5, pp. 916-931, 2015
Inner Forearm v2 dataset consists of 3 categories: Internet, IR and Standard. For each category, there are 2 sessions. The images are orginized as following:
-
Internet: 213 images in ImgSeg_01, 640 images in ImgSeg_02
-
IR: 656 images in ImgSeg_01, 656 images in ImgSeg_02
-
Standard: 656 images in ImgSeg_01, 656 images in ImgSeg_02
Image type: png
Inner Thigh v2 dataset consists of 3 categories: Internet, IR and Standard. For each category, there are 2 sessions. The images are orginized as following:
-
Internet: 229 images in ImgSeg_01, 504 images in ImgSeg_02
-
IR: 524 images in ImgSeg_01, 524 images in ImgSeg_02
-
Standard: 524 images in ImgSeg_01, 524 images in ImgSeg_02
Image type: png
Session 1:
Number of images: 100
Number of subjects: 67 (The left leg images were flipped and considered as right leg images from other subjects)
Image type: png
Session 2:
Number of images: 668
Number of subjects: 479 (The left leg images were flipped and considered as right leg images from other subjects)
Image type: png
Please cite the following paper to use the dataset:
Frodo Kin Sun Chan, Adams Wai Kin Kong, "A further study of low resolution androgenic hair patterns as a soft biometric trait", In Image and Vision Computing, 2017, ISSN 0262-8856
All the images in NTU Tattoo Dataset are collected from Flickr. The images were taken from diverse viewpoints, poses and environments with complex backgrounds, including indoor and outdoor. The raw image sizes range from 72 by 96 pixels to 500 by 500 pixels.
Number of images: 10000 (5740 tattoo images and 4260 non-tattoo images)
Image type: Color
Please cite the following paper to use the dataset:
Qingyong Xu, S. Ghosh, X. Xu, Yi Huang and A. W. K. Kong, "Tattoo detection based on CNN and remarks on the NIST database," 2016 International Conference on Biometrics (ICB), Halmstad, 2016, pp. 1-7.
To acquire NTU dataset, please first download the "Data Release Agreement.pdf" file. Open this file using pdf reader. Select the dataset you want to acquire in Section A. Print the agreement and sign on page 2. Scan the signed copy and send back to yan.ren@ntu.edu.sg with title "Application for NTU Dataset". A download link to the corresponding dataset will be send to you once after we receive the signed agreement file.
This database, a correpsonding data release agreement and project site can be found here
Please cite the following paper when using this dataset:
Wojciech Michal Matkowski, Frodo Kin Sun Chan and Adams Wai Kin Kong. "A Study on Wrist Identification for Forensic Investigation." Image and Vision Computing, vol. 88, August 2019, pp 96-112. https://doi.org/10.1016/j.imavis.2019.05.005
This database, a correpsonding data release agreement and project site can be found here
Please cite the following paper when using this dataset:
Wojciech Michal Matkowski, Krzysztof Matkowski, Adams Wai Kin Kong and Cory Lloyd Hall. "The Nipple-Areola Complex for Criminal Identification." In International Conference on Biometrics (ICB), June 2019.
This database, a correpsonding data release agreement and project site can be found here
Please cite the following paper when using this dataset:
Wojciech Michal Matkowski, Tingting Chai and Adams Wai Kin Kong. “Palmprint Recognition in Uncontrolled and Uncooperative Environment.” IEEE Transactions on Information Forensics and Security, October 2019, DOI: 10.1109/TIFS.2019.2945183.
This database, a correpsonding data release agreement and project site can be found here
Please cite the following paper when using this dataset:
Wojciech Michal Matkowski, Tingting Chai and Adams Wai Kin Kong. “Palmprint Recognition in Uncontrolled and Uncooperative Environment.” IEEE Transactions on Information Forensics and Security, October 2019, DOI: 10.1109/TIFS.2019.2945183.