Zero-crossing rate (ZCR) is one of the most important acoustic feature that has been widely used in voice activity detection, voiced/unvoiced speech classification, music/speech classification image processing, optics...
详细信息
Zero-crossing rate (ZCR) is one of the most important acoustic feature that has been widely used in voice activity detection, voiced/unvoiced speech classification, music/speech classification image processing, optics, biomedical engineering, radar and fluid mechanics. The conventional time-domain ZCR measurement is sensitive to nonstationary noise. In this paper, we present a noise robust zerocrossing rate computation method. The number of zerocrossings is computed in autocorrelation-domain rather than in time-domain. The accuracy of the proposed measurement method is evaluated using both sinusoidal and speech waveforms under different signal-to-noise ratios (SNRs). Experimental results show that the proposed ZCR measurement method achieves better accuracy than the conventional ZCR measurement methods in the literature.
This paper is an attempt to show that a dependency parser for Malayalam language can be produced using Integer Linear Programming Approach. We describe a process for developing a parser by incorporating the Paninian G...
详细信息
作者：
Prasanth, VenugopalHari Prasad, R.Soman, K.P.IV B.tech EEE
Amrita Vishwa Vidyapeetham Coimbatore TamilNadu-641105 India CEN
Centre for Excellence in Computational Engineering and Networking Amrita Vishwa Vidyapeetham Coimbatore TamilNadu-641105 India
Modernization of Indian railways has always been a question in focus for the development of the basic infrastructure of our country. Since the railways represent one of the best modes of transport available to the com...
详细信息
The principal drawback of the existing cryptographic algorithms is the maintenance of their key's secrecy. Added with, human users have a difficult time remembering strong but lengthy cryptographic keys. As a resu...
详细信息
The principal drawback of the existing cryptographic algorithms is the maintenance of their key's secrecy. Added with, human users have a difficult time remembering strong but lengthy cryptographic keys. As a result, utilizing individual's biometric features in the generation of strong and repeatable cryptographic keys has gained enormous popularity among researchers. The unpredictability of the user's biometric features, incorporated into the generated cryptographic key, makes the key unguessable to an attacker lacking noteworthy knowledge of the user's biometrics. Nevertheless, if a person's biometric is lost once, it will be compromised forever as it is inherently associated with the user. To overcome the above, cancelable biometrics has been proposed as an effective solution for canceling and re-issuing biometric templates. Here, we propose an innovative and efficient approach to generate a non-invertible cryptographic key from cancelable fingerprint templates. Initially, a one-way transformation is applied on the minutiae points extracted from the fingerprints, to attain a set of transformed points. Subsequently, the transformed points are made use of to form cancelable templates. The cancelable fingerprint templates are then utilized to generate a unique non-invertible key. As the cryptographic key generated is non-invertible, it is highly infeasible to acquire the cancelable fingerprint templates or the original fingerprint from the generated key. The effectiveness of the proposed approach is demonstrated by making use of fingerprints accessible from public sources.
The increasing immensity of identity theft in our society has resulted in the prioritization of dependable information security *** the fact that it is possible to bring about information security with the aid of a pr...
详细信息
The increasing immensity of identity theft in our society has resulted in the prioritization of dependable information security *** the fact that it is possible to bring about information security with the aid of a prevailing such as like cryptography,protecting the confidentiality of the cryptographic keys is one of the major issues to be *** is possible to effectively solve the issue by combining biometrics and ***,the improved performance of cryptographic key produced from biometrics in terms of protection has acquired massive reputation amid the researchers and ***,there is an enduring association amid the biometric and the user where in alteration is ***,the biometric is vanished eternally and probably for all the applications that apply it,when a compromise of the biometric identifier is *** might be resolved by building revocable biometric templates through cancelable ***,we have projected an approach to produce irrevocable cryptographic key from cancelable fingerprint templates,a process which proposed algorithm carries out in an efficient *** begin with,the fingerprints are utilized to extract the minutiae points that are altered,in an efficient manner so as to acquire transformed ***,the transformed points are used to produce the cancelable templates which are in turn utilized for the extraction of irrevocable *** is extremely unfeasible to obtain cancelable fingerprint templates and original fingerprints from the generated key since the cryptographic key produced is *** fingerprints accessible from open sources have been employed to assess the efficacy of our approach.
Ischemic heart disease (IHD) is one of the toughest challenges to doctors in-making right decisions due to its skimpy symptoms and complexity. We have analyzed IHD data from 65 patients to provide an aid for decision-...
详细信息
Ischemic heart disease (IHD) is one of the toughest challenges to doctors in-making right decisions due to its skimpy symptoms and complexity. We have analyzed IHD data from 65 patients to provide an aid for decision-making. Decision trees give potent structural information about the data and thereby serve as a powerful data mining tool. Support vector machines serve as excellent classifiers and predictors and can do so with high accuracy. Our tree based classifier uses non-linear proximal support vector machines (PSVM). The accuracy is very high (100% for training data) and the tree is small and precise.
暂无评论