Abstract

We consider the problem of automatic identification of faces in videos such as movies, given a dictionary of known faces from a public or an alternate database. This has applications in video indexing, content based search, surveillance, and real time recognition on wearable computers. We propose a two stage approach for this problem. First, we recognize the faces in a video using a sparse representation framework using l1-minimization and select a few key-frames based on a robust confidence measure.We then use transductive learning to propagate the labels from the key-frames to the remaining frames by incorporating constraints simultaneously in temporal and feature spaces. This is in contrast to some of the previous approaches where every test frame/track is identified independently, ignoring the correlation between the faces in video tracks. Having a few key frames belonging to few subjects for label propagation rather than a large dictionary of actors reduces the amount of confusion. We evaluate the performance of our algorithm on Movie Trailer face dataset and five movie clips, and achieve a significant improvement in labeling accuracy compared to previous approaches

Abstract

No Results Found

Abstract

Widely used online commerce systems require an user to submit his sole banking credentials or credit card details for availing desired services, thus involving high risks with untrusted service providers. Often used one-time password based systems provide additional transaction security, but are still incapable of differentiating between a genuine user trying to authenticate or an adversary with stolen credentials. This brings out a strong need for biometrics based onetime password systems. In this paper we propose a one-time biometric token based authentication protocol which works within the framework of current online transaction schemes allowing an user to carry out a financial transaction with a service provider which completes with an authorization from the bank. The proposed protocol is based on key-binding biometric cryptosystems and upholds the requirements of secure authentication, template protection and revocability while providing privacy to individual’s biometrics and anonymity from the service provider. We demonstrate our system’s security and performance using iris biometrics to authenticate individuals.

Abstract

Motion trajectories extracted from certain videos contain sufficient spatio-temporal information which can be effectively used to characterize those videos. But the task of framing text-based queries for such videos in content-based video retrieval systems is very complicated. Sketch based query is an efficient tool to construct motion-based queries but perceptual differences like spatial and temporal variability pose serious challenges to query modelling. In this work we propose a new method of modelling sketch based queries which attempts to extract the qualitative features of motion by minimizing the perceptual variability. We also develop a multilevel filter for indexing a query, in which the search results are refined at each stage using a cumulative scoring mechanism. Finally, we show the effectiveness of our algorithm on a dataset of real pool videos and a synthetic dataset containing simulated videos having very complex motion trajectories.

Abstract

We present an approach for learning low- and high-level fingerprint structures in an unsupervised manner, which we use for enhancement of fingerprint images and estimation of orientation fields, frequency images, and region masks. We incorporate the use of a convolutional deep belief network to learn features from greyscale, clean fingerprint images. We also show that reconstruction performed by the learnt network works as a suitable enhancement of the fingerprint, and hierarchical probabilistic inference is able to estimate overall fingerprint structures as well. Our approach performs better than Gabor-based enhancement and short time Fourier transform-assisted enhancement on images it was trained on. We further use information from the learnt features in first layer, which are short and oriented ridge structures, to extract the orientation field, frequency image, and region mask of input fingerprints

Abstract

Text to speech (TTS) systems hold promise as an information access tool for literate and illiterate including visually challenged. Current TTS systems can convert a typical text into a natural sounding speech. However, auditory rendering of mathematical content, specifically equation reading is not a trivial task. Mathematical equations have to be read so that appropriate bracketing such as parentheses, superscripts and subscripts are conveyed to the listener in an accurate way. Earlier works have attempted to use pauses as acoustic cues to indicate some of the semantics associated with the mathematical symbols. In this paper, we first analyse the acoustic cues which human-beings employ while speaking the mathematical content to (visually challenged) listeners and then propose four techniques which render the observed patterns in a text-tospeech system. The evaluation considered eight aspects such as listening effort, content familiarity, accentuation, intonation, etc. Our objective metrics show that a combination of the proposed techniques could render the mathematical equations using a TTS system as good as that of a human-being.

Abstract

Recently, Awasthi and Srivastava proposed a novel biometric remote user authentication scheme for the telecare medicine information system (TMIS) with nonce. Their scheme is very efficient as it is based on efficient chaotic one-way hash function and bitwise XOR operations. In this paper, we first analyze Awasthi-Srivastava’s scheme and then show that their scheme has several drawbacks: (1) incorrect password change phase, (2) fails to preserve user anonymity property, (3) fails to establish a secret session key beween a legal user and the server, (4) fails to protect strong replay attack, and (5) lacks rigorous formal security analysis. We then a propose a novel and secure biometric-based remote user authentication scheme in order to withstand the security flaw found in Awasthi-Srivastava’s scheme and enhance the features required for an idle user authentication scheme. Through the rigorous informal and formal security analysis, we show that our scheme is secure against possible known attacks. In addition, we simulate our scheme for the formal security verification using the widely-accepted AVISPA (Automated Validation of Internet Security Protocols and Applications) tool and show that our scheme is secure against passive and active attacks, including the replay and man-in-the-middle attacks. Our scheme is also efficient as compared to Awasthi-Srivastava’s scheme.

Abstract

With the rapid growth of the Internet, a lot of electronic patient records (EPRs) have been developed for e-medicine systems. The security and privacy issues of EPRs are important for the patients in order to understand how the hospitals control the use of their personal information, such as name, address, e-mail, medical records, etc. of a particular patient. Recently, Lee et al. proposed a simple group password-based authenticated key agreement protocol for the integrated EPR information system (SGPAKE). However, in this paper, we show that Lee et al.’s protocol is vulnerable to the off-line weak password guessing attack and as a result, their scheme does not provide users’ privacy. To withstand this security weakness found in Lee et al.’s scheme, we aim to propose an effective dynamic group password-based authenticated key exchange scheme for the integrated EPR information system, which retains the original merits of Lee et al.’s scheme. Through the informal and formal security analysis, we show that our scheme provides users’ privacy, perfect forward security and known-key security, and also protects online and offline password guessing attacks. Furthermore, our scheme efficiently supports the dynamic group password-based authenticated key agreement for the integrated EPR information system. In addition, we simulate our scheme for the formal security verification using the widely-accepted AVISPA (Automated Validation of Internet Security Protocols and Applications) tool and show that our scheme is secure against passive and active attacks.

Abstract

Advancement in communication technology provides a scalable platform for various services, where a remote user can access the server from anywhere without moving from its place. It provides a unique opportunity for online services such that a user does not need to be physically present at the service center. These services adopt authentication and key agreement protocols in order to ensure authorized and secure access to the resources. Most of the authentication schemes proposed in the literature support a single-server environment, where the user has to register with each server. If a user wishes to access multiple application servers, he/she requires to register with each server. The multi-server authentication introduces a scalable platform such that a user can interact with any server using single registration. Recently, Chuang and Chen proposed an efficient multi-server authenticated key agreement scheme based on a user’s password and biometrics (Chuang and Chen, 2014). Their scheme is a lightweight, which requires the computation of only hash functions. In this paper, we first analyze Chuang and Chen’s scheme and then identify that their scheme does not resist stolen smart card attack which causes the user’s impersonation attack and server spoofing attack. We also show that their scheme fails to protect denial-of-service attack. We aim to propose an efficient improvement on Chuang and Chen’s scheme to overcome the weaknesses of their scheme, while also retaining the original merits of their scheme. Through the rigorous informal and formal security analysis, we show that our scheme is secure against various known attacks including the attacks found in Chuang and Chen’s scheme. Furthermore, we simulate our scheme for the formal security verification using the widely-accepted AVISPA (Automated Validation of Internet Security Protocols and Applications) tool and show that our scheme is secure against the replay and man-in-the-middle attacks. In addition, our scheme is comparable in terms of the communication and computational overheads with Chuang and Chen’s scheme and other related existing schemes.

Abstract

A user authentication in the distributed computer networks (DCNs) plays a crucial rule to verify whether the user is a legal user and can therefore be granted access to the requested services to that user. In recent years, several RSA‐based single sign‐on mechanisms have been proposed in DCNs. However, most of them cannot preserve the user anonymity when possible attacks occur. The user devices are usually battery limited (e.g., cellular phones) and the elliptic‐curve cryptosystem is much efficient than RSA cryptosystem for the battery‐limited devices. In this paper, we aim to propose a new secure elliptic‐curve cryptosystem‐based single sign‐on mechanism for user authentication and key establishment for the secure communications in a DCNs using biometric‐based smart card. In our scheme, a user only needs to remember a private password and his or her selected unique identity to authenticate and agree on a high‐entropy cryptographic one‐time session key with a provider to communicate over untrusted public networks. Through formal and informal security analysis, we show that our scheme prevents other known possible attacks. In addition, we perform simulation on our scheme for the formal security verification using the widely‐accepted Automated Validation of Internet Security Protocols and Applications tool. The simulation results ensure that our scheme is secure against replay and man‐in‐the‐middle attacks. Furthermore, our scheme provides high security along with lower computational cost and communication cost, and as a result, our scheme is much suitable for the battery‐limited devices as compared to other related RSA‐based schemes.