Jeung-Yoon Choi

Alignment-Free Cancelable Fingerprint Templates Based on Local Minutiae Information

To replace compromised biometric templates, cancelable biometrics has recently been introduced. The concept is to transform a biometric signal or feature into a new one for enrollment and matching. For making cancelable fingerprint templates, previous approaches used either the relative position of a minutia to a core point or the absolute position of a minutia in a given fingerprint image. Thus, a query fingerprint is required to be accurately aligned to the enrolled fingerprint in order to obtain identically transformed minutiae. In this paper, we propose a new method for making cancelable fingerprint templates that do not require alignment. For each minutia, a rotation and translation invariant value is computed from the orientation information of neighboring local regions around the minutia. The invariant value is used as the input to two changing functions that output two values for the translational and rotational movements of the original minutia, respectively, in the cancelable template. When a template is compromised, it is replaced by a new one generated by different changing functions. Our approach preserves the original geometric relationships (translation and rotation) between the enrolled and query templates after they are transformed. Therefore, the transformed templates can be used to verify a person without requiring alignment of the input fingerprint images. In our experiments, we evaluated the proposed method in terms of two criteria: performance and changeability. When evaluating the performance, we examined how verification accuracy varied as the transformed templates were used for matching. When evaluating the changeability, we measured the dissimilarities between the original and transformed templates, and between two differently transformed templates, which were obtained from the same original fingerprint. The experimental results show that the two criteria mutually affect each other and can be controlled by varying the control parameters of the changing functions.

Changeable Biometrics for Appearance Based Face Recognition

To enhance security and privacy in biometrics, changeable (or cancelable) biometrics have recently been introduced. The idea is to transform a biometric signal or feature into a new one for enrollment and matching. In this paper, we proposed changeable biometrics for face recognition using an appearance based approach. PCA and ICA coefficient vectors extracted from an input face image are normalized using their norm. The two normalized vectors are scrambled randomly and a new transformed face coefficient vector (transformed template) is generated by addition of the two normalized vectors. When a transformed template is compromised, it is replaced by using a new scrambling rule. Because the transformed template is generated by the addition of two vectors, the original PCA and ICA coefficients cannot be recovered from the transformed coefficients. In our experiment, we compared the performance between the cases when PCA and ICA coefficient vectors are used for verification and when the transformed coefficient vectors are used for verification.

Detection of consonant voicing: a module for a hierarchical speech recognition system.

This research describes a module for detecting consonant voicing in a hierarchical speech recognition system. In this system, acoustic cues are used to infer values of features that describe phonetic segments. A first step in the process is examining consonant production and conditions for phonation, to find acoustic properties that may be used to infer consonant voicing. These are examined in different environments to determine a set of reliable acoustic cues. These acoustic cues include fundamental frequency, difference in amplitudes of the first two harmonics, cutoff first formant frequency, and residual amplitude of the first harmonic, around consonant landmarks. Classification experiments are conducted on hand and automatic measurements of these acoustic cues for isolated and continuous speech utterances. Voicing decisions are obtained for each consonant landmark, and are compared with lexical and perceived voicing for the consonant. Performance is found to improve when measurements at the closure and release are combined. Training on isolated utterances gives classification results for continuous speech that is comparable to training on continuous speech. The results in this study suggest that acoustic cues selected by considering the representation and production of speech may provide reliable criteria for determining consonant voicing.

Using Genetic Algorithms to Improve Matching Performance of Changeable biometrics from Combining PCA and ICA Methods

Biometrics is personal authentication which uses an individuals information. In terms of user authentication, biometric systems have many advantages. However, despite its advantages, they also have some disadvantages in the area of privacy problems. Changeable biometrics is solution to problem of privacy protection. In this paper we propose a changeable face biometrics system to overcome this problem. The proposed method uses the PCA and ICA methods and genetic algorithms. PCA and ICA coefficient vectors extracted from an input face image were normalized using their norm. The two normalized vectors were transformed using a weighting matrix which is derived using genetic algorithms and then scrambled randomly. A new transformed face coefficient vector was generated by addition of the two weighted normalized vectors. Through experiments, we see that we can achieve performance accuracy that is better than conventional methods. And, it is also shown that the changeable templates are non-invertible and provide sufficient reproducibility.

HMM with global path constraint in Viterbi decoding for isolated word recognition

Hidden Markov models (HMMs) with explicit state duration density (HMM/SD) can represent the time-varying characteristics of speech signals more accurately. However, such an advantage is reduced in relatively smooth state duration densities or long bounded duration. To solve this problem, the authors propose HMMs with global path constraint (HMM/GPC) where the transition between states occur only within prescribed time slots. HMM/GPC explicitly limits state durations and accurately describes the temporal structure of speech simply and efficiently. HMMs formed by combining HMM/GPC with HMM/SD are also presented (HMM/SD+GPC) and performances are compared. HMM/GPC can be implemented with slight modifications to the conventional Viterbi algorithm. HMM/GPC and HMM/SD+GPC not only show superior performance than the conventional HMM and HMM/SD but also require much less computation.<<ETX>>

Perceptual relevance of the temporal envelope to the speech signal in the 4–7kHz band

The perceptual relevance of adopting the temporal envelope to model the frequency band of 4-7 kHz (highband) in wideband speech signal is described in this letter. Based on theoretical work in psychoacoustics, we find out that the temporal envelope can indeed be a perceptual cue for the high-band signal, i.e., a noiseless sound can be obtained if the temporal envelope is roughly preserved. Subjective listening tests verify that transparent quality can be obtained if the model is used for the 4.5-7 kHz band. The proposed model has the benefits of offering flexible scalability and reducing the cost for quantization in coding applications.

Pause length and speech rate as durational cues for prosody markers

Two durational cues for prosody are examined for utterances in the Radio Speech Corpus. The acoustic cues are length of pause and speech rate associated with each syllabic segment. Pauses are defined as intervals where probability of voicing and energy (rms) fall below respective thresholds. Speech rate is defined as the reciprocal of the duration of the syllabic segment, without normalization with respect to speaker or segment identity. Each syllabic segment was assigned prosodic markers that are a combination of degree of accent and boundary, i.e., {unaccented, accented} × {nonboundary, intermediate boundary, intonational boundary}. Distributions of pause length and speech rate for each type of prosodic marker show that in general, mean pause length increases and speech rate decreases as the strength of the boundary increases from nonboundary to intonational boundary. Accented syllabic segments at intermediate or intonational boundaries showed longer associated pause length and slower speech rate than unaccented segments. Also, for different types of intonational boundary tones, mean pause length decreases in the order of: L‐L%, H‐H%, H‐L%, L‐H%. Preliminary classification results using the two durational cues show detection rates for intonational boundaries around 75%, with insertion rates around 25%.

Phonetically optimized speaker modeling for robust speaker recognition

This paper proposes an efficient method to improve speaker recognition performance by dynamically controlling the ratio of phoneme class information. It utilizes the fact that each phoneme contains different amounts of speaker discriminative information that can be measured by mutual information. After classifying phonemes into five classes, the optimal ratio of each class in both training and testing processes is adjusted using a non-linear optimization technique, i.e., the Nelder-Mead method. Speaker identification results verify that the proposed method achieves 18% improvement in terms of error rate compared to a baseline system.

Analysis of acoustic parameters for consonant voicing classification in clean and telephone speech

This paper describes acoustic cues for classification of consonant voicing in a distinctive feature-based speech recognition system. Initial acoustic cues are selected by studying consonant production mechanisms. Spectral representations, band-limited energies, and correlation values, along with Mel-frequency cepstral coefficients features (MFCCs) are also examined. Analysis of variance is performed to assess relative significance of features. Overall, 82.2%, 80.6%, and 78.4% classification rates are obtained on the TIMIT database for stops, fricatives, and affricates, respectively. Combining acoustic parameters with MFCCs shows performance improvement in all cases. Also, performance in the NTIMIT telephone channel speech shows that acoustic parameters are more robust than MFCCs.

Detection of obstruent consonant landmark for knowledge based speech recgonition system

Obstruent consonant landmarks are detected using spectral energy difference profiles. This study expands upon previous work by Liu. A[J. Acoust. Soc. Am. 100, 3417‐3430, 1996]. The proposed algorithm detects four types of landmarks : [stop closure], [stop release], [fricative closure] and [fricative release], where affricates are detected by combining [stop closure], [fricative closure] and [fricative release]. In addition to finding abrupt changes in energy differences, we use energy contours, relative energy and spectral center of gravity differences. This method results in improved performance particularly for CV obstruents. Overall detection rates for stop closure and release are 76.9% and 85.7% for obstruent landmarks in TIMIT, and fricatives yield 82.2% and 83.6% respectively. For strident fricatives, the figures are 94.7% and 93.6%.