NIST SD27: Annotations & Results

This page provides some annotations and results of NIST SD 27 for studying the problem of feature extraction of latent fingerprints. The annotations are manually labeled features, including orientation fields, period maps, skeletons, singular points and region of interests. The results are orientation fields estimated by algorithms of GlobalDict[1] and LocalDict[2], as well as poses of latent fingerprints estimated by LocalDict[2]. Some annotations are shown below:


Orientation fields: Blockwise orientation elements from -90 degrees to 90 degrees. The block size is 16 pixels. 91 degrees indicate background.

Minutiae: [x,y,theta,quality,type]

Singular points:[x,y,type]

Minutiae pairs: m*_1.txt & m*_2.txt are minutiae extracted by COTS from file and latent fingerprint, respectively. P*-*.txt are manually marked pairs, each line [i,j] means the ith minutia in m*_1.txt and the jth minutia in m*_2.txt are mated.

Pose: [x,y,theta]


You are free to download the dataset for non-commercial research and educational purposes.

Annotations: link link link link link link

Results: link link


Research papers that used this database should cite the following paper:

[1] J. Feng, J. Zhou, and A. K. Jain, “Orientation field estimation for latent fingerprint enhancement,” IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 35, no. 4,pp. 925–40, 2013.

[2] X. Yang, J. Feng, and J. Zhou, “Localized dictionaries based orientation field estimation for latent fingerprints,” IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 36, no. 5, pp. 955–969, 2014.


Contact Dr. Jianjiang Feng for further information.