Fingerprints are commonly used biometric trait used for identification. Latent prints are the fingerprint impressions which are inadvertently left by a person on different surfaces that come in contact with the finger at the crime scene. These latent fingerprints are used as evidence in the forensics to identify the suspect. Sometimes one fingerprint gets overlapped on another fingerprint, due to which it becomes difficult to extract features from the fingerprint and identify the suspect. Till now, overlapped and non-overlapped regions were segmented manually, which need extra human effort and consumes a lot of time. So, separating overlapped fingerprint automatically is necessary to identify the correct person. In this paper, machine learning algorithm is used to segment the overlapped fingerprint regions automatically by extracting features using Random Decision Forest (RDF) classifier and then separating the two fingerprints. Also, a database of overlapped latent fingerprints is collected using the touch less based sensor called Reflected Ultra Violet Imaging System (RUVIS). This device is used to search, view, detect and capture the latent fingerprints on non-porous surfaces. The performance of the proposed approach is evaluated on the developed database by computing False Rejection Rate (FRR).

Biometrics, latent fingerprints, Overlapped fingerprints, Random Decision Forest, Reflected Ultra Violet Imaging System

[1] F. Chen, J. Feng, A. K. Jain, J. Zhou, and J. Zhang, “Separating overlapped fingerprints,” IEEE Trans. Inf. Foren. Secur., Vol 76, Issue 10, pp.346–359, 2011.
[2] Y. Shi, J. Feng, J. Zhou, “Separating overlapped fingerprints using constrained relaxation labeling”, In: Proceedings of the 2011 international joint conference on biometrics, 2011.
[3] J. Feng, Y. Shi, J. Zhou, “Robust and efficient algorithms for separating latent overlapped fingerprints”, IEEE Trans Inf Forensics Secur Vol. 7, Issue 5, pp.1498–1510, 2012.
[4] Q. Zhao, A. Jain, “Model based separation of overlapping latent fingerprints”, IEEE Trans Inf Forensics Secur Vol.7, Issue 3, pp.904–918, 2012.
[5] N. Zhang, Y. Zang, X. Yang, X. Jia, J. Tian, “Adaptive orientation model fitting for latent overlapped fingerprints separation”, IEEE Trans Inf Forensics Secur, Vol 9, Issue 10, pp.1547–1556, 2014.
[6] S.Jeyanthi, N.U. Maheswari and R. Venkatesh. "Neural network based automatic fingerprint recognition system for overlapped latent images." Journal of Intelligent & Fuzzy Systems, Vol 28, Issue 6, pp.2889-2899, 2015.
[7] S. Jeyanthi, N.U. Maheswari and R. Venkatesh. "An Efficient Automatic Overlapped Fingerprint Identification and Recognition Using ANFIS Classifier", International Journal of Fuzzy Systems, Vol 18, Issue 3, pp.478-491, 2015.
[8] B. Stojanović, A. Nešković, O. Marques, “A novel neural network based approach to latent overlapped fingerprints separation”, Multimedia Tools and Applications, Vol 76, Issue 10, pp.12775–12799, 2017.
[9] B. Stojanović, O. Marques, A. Nešković, S. Puzović, "Fingerprint ROI segmentation based on deep learning", Telecommunications Forum (TELFOR), IEEE, Volume 76, Issue 10, pp.1-4, 2016.
[10] A. Sankaran, A. Jain, T. Vashisth, M. Vatsa, R. Singh, "Adaptive latent fingerprint segmentation using feature selection and random decision forest classification", Information Fusion, Elsevier, Volume 34, Issue 10, pp.1-15, 2017.
[11] K. Tejas, C. Swathi, D.A. Kumar and R. Muthu, "Automated region masking of latent overlapped fingerprints", In Power and Advanced Computing Technologies (i-PACT), IEEE, Innovations in pp.1-6, 2017.
[12] S.U. Maheswari, and E. Chandra. "An Enhanced Active contour based Segmentation for Fingerprint Extraction." International Journal on Computer Science and Engineering, Vol 4, Issue 9, pp.1633, 2012.
[13] K. Qian, M. Schott and J. Dittmann, “Separation of contactless captured high-resolution overlapped latent fingerprints: parameter optimisation and evaluation”, In Biometrics and Forensics (IWBF), International Workshop, IEEE, pp.1-4, 2013.
[14] K. Qian, M. Schott, W. Zheng and J. Dittmann, “Context-based approach of separating contactless captured high-resolution overlapped latent fingerprints”, IET biometrics, Vol. 3, Issue 2, pp,101-112, 2014.
[15] S. Jeyanthi, N.U. Maheswari and R. Venkatesh, “Separation and recognition of overlapped latent images”, In Computing, Communications and Networking Technologies (ICCCNT), IEEE, Fourth International Conference on pp.1-6, 2013.