Fingerprint singular points extraction based on orientation tensor field and Laurent series

Singular point(SP)extraction is a key component in automatic fingerprint identification system(AFIS).A new method was proposed for fingerprint singular points extraction,based on orientation tensor field and Laurent series.First,fingerprint orientation flow field was obtained,using the gradient of fingerprint image.With these gradients,fingerprint orientation tensor field was calculated.Then,candidate SPs were detected by the cross-correlation energy in multi-scale Gaussian space.The energy was calculated between fingerprint orientation tensor field and Laurent polynomial model.As a global descriptor,the Laurent polynomial coefficients were allowed for rotational invariance.Furthermore,a support vector machine(SVM)classifier was trained to remove spurious SPs,using cross-correlation coefficient as a feature vector.Finally,experiments were performed on Singular Point Detection Competition 2010(SPD2010)database.Compared to the winner algorithm of SPD2010 which has best accuracy of 31.90%,the accuracy of proposed algorithm is 45.34%.The results show that the proposed method outperforms the state-of-the-art detection algorithms by large margin,and the detection is invariant to rotational transformations.

Robust Fingerprint Localization Based on Massive MIMO Systems

Location-based services have become an important part of the daily life.Fingerprint localization has been put forward to overcome the shortcomings of the traditional positioning algorithms in indoor scenario and rich scattering environment.In this paper,a single-site multiple-input multiple-output(MIMO)orthogonal frequency division multiplexing(OFDM)system is modeled,from which an angle delay channel power matrix(ADCPM)is extracted.Considering the changing environment,auto encoders are used to generate new fingerprints based on ADCPM fingerprints to improve the robustness of the fingerprints.When the scattering environment has changed beyond a certain extent,the robustness will not be able to make up for the positioning error.Under this circumstance,an updating of the fingerprint database is imperative.A new fingerprint database updating algorithm which combines a new clustering method and an updating rule based on probability is proposed.Simulation results show the desirable performance of the proposed methods.

Indoor Localization with a Crowdsourcing Based Fingerprints Collecting

Fingerprint matching is adopted by a large family of indoor localization schemes,where collecting fingerprints is inevitable but all consuming.While the increasingly popular crowdsourcing based approach provides an opportunity to relieve the burden of fingerprints collecting,a number of formidable challenges for such an approach have yet been studied.For instance,querying in a large fingerprints database for matching process takes a lot of time and calculation;fingerprints collected by crowdsourcing lacks of robustness because of heterogeneous devices problem.Those are important challenges which impede practical deployment of the fingerprint matching indoor localization system.In this study,targeting on effectively utilizing and mining large amount fingerprint data,enhancing the robustness of fingerprints under heterogeneous devices' collection and realizing the real time localization response,we propose a crowdsourcing based fingerprints collecting mechanism for indoor localization systems.With the proposed approach,massive raw fingerprints will be divided into small clusters while diverse devices' uploaded fingerprints will be merged for overcoming device heterogeneity,both of which will contribute to reduce response time.We also build a mobile cloud testbed to verify the proposed scheme.Comprehensive real world experiment results indicate that the scheme can provide comparable localization accuracy.