Marcos Faundez-Zanuy, Jiří Mekyska and Xavier Font-Aragones
Access to the database
Please fill in a license agreement that can be downloaded in DOCX or PDF and send it to email@example.com with CC to firstname.lastname@example.org. You will consequently get an access to the database.
Publications to be cited
M. Faundez-Zanuy, J. Mekyska and X. Font-Aragones, „A new hand image database simultaneously acquired in visible, near-infrared and thermal spectrums“, Cognitive Computation, vol. 6, no. 2, pp. 230–240, 2014.
X. Font-Aragones, M. Faundez-Zanuy and J. Mekyska, „Thermal hand image segmentation for biometric recognition“, IEEE Aero El Sys Mag, vol. 28, no. 6, pp. 4–14, 2013.
This database is freely available. In case of any publications based on this database please cite the papers mentioned above.
The acquisition over the VIS and TIR data was performed by a commercial thermal camera Testo 882-3. We have used a second external camera to obtain the NIR data. In this case we have built a NIR camera using a webcam changing the default optical filter for a couple of Kodak filters for IR. We have also used a printed circuit board with 16 infra-red LEDs that provide the infra-red illumination.
In order to alleviate the variability on the way the users present their hand we have used a kind of removable mask/template. Users had to put their hand in a neoprene surface with the help of a hand mask. Once the hand is placed the mask was removed and the three images (VIS, NIR, TIR) were shot. The same process was repeated with the palmar hand side, but using the mask in the opposite position (flip up to down).
Once the first acquisition was finished (no more than a minute) the user performed some exercise in order to change hand heat conditions. This step was carried out in less than 30 seconds so that when finished, the user proceeded again with the second acquisition. For each session a total of 12 hand images were captured per user.
The multispectral hand image database consists of 100 users. Each user has been acquired in 5 different recording sessions. In each session two acquisitions were performed of each hand (palm and dorsal image). The time span for the whole database acquisition was 6 months. From January to June users were acquired in five sessions. These sessions were separated from each other with a delay of at least 4 weeks. Any time a user proceed with a session two acquisitions were done. The first one was done without any special requirement. The second one was done just after the first one but with a slightly small hand exercise (flexing and stretching hand pressing a soft dice). This hand exercise helped to change hand thermal conditions.
The total number of images is:
100 users x 5 session x 2 acquisitions x 3 spectral ranges x 2 sides (palm and dorsal images) = 6.000 images.