• MALSORI
• /
• no.63
• /
• pp.125-137
• /
• 2007

Recently, many techniques have been proposed to improve the noise robustness for speaker verification. In this paper, we consider the feature recombination technique in multi-band approach. In the conventional feature recombination for speaker verification, to compute the likelihoods of speaker models or universal background model, whole feature components are used. This computation method is not effective in a view point of multi-band approach. To deal with non-effectiveness of the conventional feature recombination technique, we introduce a subband likelihood computation, and propose a modified feature recombination using subband likelihoods. In decision step of speaker verification system in noise environments, a few very low likelihood scores of a speaker model or universal background model cause speaker verification system to make wrong decision. To overcome this problem, a reliable feature selection method is proposed. The low likelihood scores of unreliable feature are substituted by likelihood scores of the adaptive noise model. In here, this adaptive noise model is estimated by maximum a posteriori adaptation technique using noise features directly obtained from noisy test speech. The proposed method using subband-based reliable feature selection obtains better performance than conventional feature recombination system. The error reduction rate is more than 31 % compared with the feature recombination-based speaker verification system.

• ETRI Journal
• /
• v.30 no.1
• /
• pp.89-100
• /
• 2008

We consider the feature recombination technique in a multiband approach to speaker identification and verification. To overcome the ineffectiveness of conventional feature recombination in broadband noisy environments, we propose a new subband feature recombination which uses subband likelihoods and a subband reliable-feature selection technique with an adaptive noise model. In the decision step of speaker recognition, a few very low unreliable feature likelihood scores can cause a speaker recognition system to make an incorrect decision. To overcome this problem, reliable-feature selection adjusts the likelihood scores of an unreliable feature by comparison with those of an adaptive noise model, which is estimated by the maximum a posteriori adaptation technique using noise features directly obtained from noisy test speech. To evaluate the effectiveness of the proposed methods in noisy environments, we use the TIMIT database and the NTIMIT database, which is the corresponding telephone version of TIMIT database. The proposed subband feature recombination with subband reliable-feature selection achieves better performance than the conventional feature recombination system with reliable-feature selection.

• Proceedings of the KSPS conference
• /
• 2007.05a
• /
• pp.127-130
• /
• 2007

The conventional feature recombination technique is very effective in the band-limited noise condition, but in broad-band noise condition, the conventional feature recombination technique does not produce notable performance improvement compared with the full-band system. To cope with this drawback, we introduce a new technique of sub-band likelihood computation in the feature recombination, and propose a new feature recombination method by using this sub-band likelihood computation. Furthermore, the reliable sub-band selection based on the signal-to-noise ratio is used to improve the performance of this proposed feature recombination. Experimental results shows that the average error reduction rate in various noise condition is more than 27% compared with the conventional full-band speaker identification system.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.279-284
• /
• 2009

Speaker verification determines whether the claimed speaker is accepted based on the score of the test utterance. In recent years, methods based on Gaussian mixture models and universal background model have been the dominant approaches for text-independent speaker verification. These speaker verification systems based on these methods provide very good performance under laboratory conditions. However, in real situations, the performance of speaker verification system is degraded dramatically. For overcoming this performance degradation, the feature recombination method was proposed, but this method had a drawback that whole sub-band feature vectors are used to compute the likelihood scores. To deal with this drawback, a modified feature recombination method which can use each sub-band likelihood score independently was proposed in our previous research. In this paper, we propose a sub-band weighting method based on sub-band signal-to-noise ratio which is combined with previously proposed modified feature recombination. This proposed method reduces errors by 28% compared with the conventional feature recombination method.

• ETRI Journal
• /
• v.33 no.2
• /
• pp.240-250
• /
• 2011

To enable a relevance feedback paradigm to evolve itself by users' feedback, a reinforcement learning method is proposed. The feature space of the medical images is partitioned into positive and negative hypercubes by the system. Each hypercube constitutes an individual in a genetic algorithm infrastructure. The rules take recombination and mutation operators to make new rules for better exploring the feature space. The effectiveness of the rules is checked by a scoring method by which the ineffective rules will be omitted gradually and the effective ones survive. Our experiments on a set of 10,004 images from the IRMA database show that the proposed approach can better describe the semantic content of images for image retrieval with respect to other existing approaches in the literature.

• Journal of the Korean Society of Costume
• /
• v.61 no.1
• /
• pp.69-83
• /
• 2011

This study applied cartoon-like simple and implicit symbol of Keith Haring works to tubular jacquard knit, designed woman casual knit wear with fun, explicit and activities using fluorescent yarns with explicit features. This study aimed to propose the directions for new market pioneering that meets the emotional requirements of modern people by raising the level of utilization and expression area of knit fashion, and to present the constructive directions for high value-added knit wear by designing highly effective knit wear. Methods of this study examined the lifetime and artistic background of Keith Haring and collected his works through searching of internet, scholastic publications or thesis. Formative types and colors of works collected were analyzed and classified according to characteristic standards of Haring and characteristic formativeness and color of works were extracted. The result of this study is as shown below: First, the feature of his works can be summarized into motif that has symbolic and implicit cartoon-like lines including narratives as well as enormous number of works. Such feature is very suitable for variation and recombination in realization of design and has unlimited potentials for development. Second, image colors can be extracted such as black and white, original color and fluorescent paints, which reflects the intention of artist who put focus on communication with the public. Such color feature is very suitable for design motif that has explicitness of reflecting the intention of artist through use of special fluorescent yarn. Third, I made it as reversible wear that has effects of increasing or decreasing the explicitness according to color area ratio using the feature of textures allowing use of both sides as well as form stability of tubular jacquard. Fourth, by adjusting the thickness of fabric with controlling of ply in fabric using poly yarn and wool union yarn, I could obtain good results of study in terms of technique that can express various materials with embossed dynamic effects of unevenness on flat surface.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.300.2-300.2
• /
• 2013

Periodically shaped pillar-arrays and hole-arrays were fabricated on a Si wafer. Geometric features are similar in a periodic length of 4 ${\mu}m$ and a depth of 2 ${\mu}m$. For the hole-array patterns, positive PR processes were performed. UV exposed PR patterns were removed during a developing process to leave shapes of inversely replicated from a glass photomask. Meanwhile, negative PR processes were taken for the pillar-array patterns. UV exposed PR patterns were remained on a Si substrate having a same feature of patterns of a glass photomask. For an electrical aspect, a pillar structure has a short carrier-collection length resulting in the improved open-circuit voltage of 609 mV from 587 mV of a planar device. An improved performance may be achieved to reduce recombination loss along the patterning surface.

• Bulletin of the Korean Space Science Society
• /
• 2010.04a
• /
• pp.29.4-30
• /
• 2010

A VHF meteor radar at King Sejong Station ($162.22^{\circ}S$, $58.78^{\circ}W$), Antarctica has been observing meteors during a period of March 2007-July 2009. We analyzed the height profiles of the observed meteor decay times between 70 and 95 km by classifying strong and weak meteors according to their estimated electron line densities. The height profiles of monthly averaged decay times show a peak whose altitude varies with season in the range of 80~85 km: higher peak in southern spring and summer than in fall and winter. The higher peak during summer is consistent with colder temperatures that cause faster chemical reactions of electron removal, as effective recombination rates measured by rocket experiments. The height profiles of 15-min averaged decay times show a clear increasing trend with decreasing altitude from 95 km to the peak altitude, especially for weak meteors. This feature for weak meteors is well explained by ambipolar diffusion of meteor trails, allowing one to estimate atmospheric temperatures and pressures, as in previous studies. However, the strong meteors show not only significant scatters but also different slope of the increasing trend from 95 km to the peak altitude. Therefore, atmospheric temperature estimation from meteor decay times should be applied for weak meteors only. In this study, we present the simple model decay times to explain the height profiles of the observed decay times and discuss the additional removal processes of meteor trail electrons through the empirical recombination and by icy particles.

• Journal of the Korean Ceramic Society
• /
• v.55 no.3
• /
• pp.244-260
• /
• 2018

Issues in the electrical characterization of semiconducting photoanodes in a photoelectrochemical (PEC) cell, such as the cell geometry dependence, scan rate dependence in DC measurements, and the frequency dependence in AC measurements, are addressed, using the example of a $TiO_2$ photoanode. Contrary to conventional constant phase element (CPE) modeling, the capacitive behavior associated with Mott-Schottky (MS) response was successfully modeled by a Havriliak-Negami (HN) capacitance function-which allowed the determination of frequency-independent Schottky capacitance parameters to be explained by a trapping mechanism. Additional polarization can be successfully described by the parallel connection of a Bisquert transmission line (TL) model for the diffusion-recombination process in the nanostructured $TiO_2$ electrode. Instead of shunt CPEs generally employed for the non-ideal TL feature, TL models with ideal shunt capacitors can describe the experimental data in the presence of an infinite-length Warburg element as internal interfacial impedance - a characteristic suggested to be a generic feature of many electrochemical cells. Fully parametrized impedance spectra finally allow in-depth physicochemical interpretations.

• Journal of the Korean Chemical Society
• /
• v.20 no.3
• /
• pp.193-197
• /
• 1976

The ultrasonic absorption has been measured in aqueous solution of dodecyl pyridinium chloride (DPC) at $20^{\circ}C$ over the frequency range of 0.1${\sim}$90 Mc.The excess absorption was observed only in solutions in which the concentration was higher than the critical micellar concentration(cmc). The mechanism for this feature was attributed to the reaction $M_2\;{\rightleftharpoons}\;M_1\;＋\;1.2Cl^-$ Where $M_2$ and $M_1$ and M1 are two types of micelle. The rate constants of the forward and the reverse reactions were $6.6{\times}10^5 sec^{-1}\;and\;2.7{\times}10^11sec^{-1}mol^{-1.2}$respectively. Some kinetic charateristics including the free energy, enthalpy and entropy were calculated.