• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.115-118
• /
• 2002

Parameters used in a speaker recognition system are desirable expressing speaker's characteristics filly and have in a speech. That is to say, if inter-speaker than intra-speaker variance a big characteristic, it is useful to distinguish between speakers. Also, to make minimum error between speakers, it is required the improved recognition technology as well as the distinguishing characteristics. When we see the result of recent simulation performance, we obtain more exact performance by using dynamic characteristics and constant characteristics by a speaking habit. Therefore we suggest it to solve this problem as followings. The prosodic information is used by a characteristic vector of speech. Characteristics vector generally using in speaker recognition system is a modeling spectrum information and is working for a high performance in non-noise circumstance. However, it is found a problem that characteristic vector is distorted in noise circumstance and it makes a reduction of recognition rate. In this paper, we change pitch line divided by segment which can estimate a dynamic characteristic and it is used as a recognition characteristic. we confirmed that the dynamic characteristic is very robust in noise circumstance with a simulation. We make a decision of acceptance or rejection by comparing test pattern and recognition rate using the proposed algorithm has more improvement than using spectrum and prosodic information. Especially stational recognition rate can be obtained in noise circumstance through the simulation.

• Applied Chemistry for Engineering
• /
• v.16 no.2
• /
• pp.257-261
• /
• 2005

The microbial cellulose is in a form of three dimensional net structures that consists of 20~50 nm fibrils. It possesses high crystallinity and orientation. It is difficult to synthesize large amount of fibrous carbon nanomaterials by the carbonization process using raw materials such as polyacrylonitrile (PAN), regenerated cellulose (Rayon) and pitch. However, it seems possible thru the application of microbial cellulose as raw material. The application of such cellulose can be further extended to the synthesis of highly oriented graphite fiber. Out of three different cellulose-producing strains, G. xylinus ATCC11142 was chosen as it has the highest productivity (0.066 g dried cellulose/15 mL medium). Tar is often produced during the carbonization of cellulose that limits the formation fibrous structure of the carbonized sample. In order to solve such a problem, pre-studied purification methods of carbon nanotube such as liquid phase oxidation, gas phase oxidation and filtration associated with ultrasonication were applied at the carbonized cellulose. In that case. only by filtration associated with ultrasonication, improved the formation of fiber structure of the carbonized cellulose.

• Journal of the Microelectronics and Packaging Society
• /
• v.6 no.2
• /
• pp.37-43
• /
• 1999

Modules of the system that requires large capacity and high-speed information processing are implemented in the form of MCM that allows high-speed data processing, high density circuit integration and widely applied to such fields as ATM, GPS and PCS. Hence we developed the ATM switching module that is consisted of three chips and 2.48 Gbps data throughput, in the form of 10 multi-layer by Cu/Photo-BCB and 491pin PBGA which size is $48 \times 48 \textrm {mm}^2$. hnologies required for the development of the MCM includes extracting parameters for designing the substrate/package through the interconnect characterization to implement the high-speed characteristics, thermal management at the high-density MCM, and the generation of the testability that is one of the most difficult issues for developing the MCM. For the development of the ATM Switching MCM, we extracted signaling delay, via characteristics and crosstalk parameters through the interconnect characterization on the MCM-D. For the thermal management of 15.6 Watt under the high-density structure, we carried out the thermal analysis. formed 1.108 thermal vias through the substrate, and performed heat-proofing processing for the entire package so that it can keep the temperature less than $85^{\circ}C$. Lastly, in order to ensure the testability, we verified the substrate through fine pitch probing and applied the Boundary Scan Test (BST) for verifying the complex packaging/assembling processes, through which we developed an efficient and cost-effective product.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.411-411
• /
• 2012

Over the recent years, surface enhanced Raman spectroscopy (SERS) has dramatically grown as a label-free detecting technique with the high level of selectivity and sensitivity. Conventional SERS-active nanostructured layers have been deposited or patterned on rigid substrates such as silicon wafers and glass slides. Such devices fabricated on a flexible platform may offer additional functionalities and potential applications. For example, flexible SERS-active substrates can be integrated into microfluidic diagnostic devices with round-shaped micro-channel, which has large surface area compared to the area of flat SERS-active substrates so that we may anticipate high sensitivity in a conformable device form. We demonstrate fabrication of flexible SERS-active nanostructured substrates based on soft-lithography for simple, low-cost processing. The SERS-active nanostructured substrates are fabricated using conventional Si fabrication process and inkjet printing methods. A Si mold is patterned by photolithography with an average height of 700 nm and an average pitch of 200 nm. Polydimethylsiloxane (PDMS), a mixture of Sylgard 184 elastomer and curing agnet (wt/wt = 10:1), is poured onto the mold that is coated with trichlorosilane for separating the PDMS easily from the mold. Then, the nano-pattern is transferred to the thin PDMS substrates. The soft lithographic methods enable the SERS-active nanostructured substrates to be repeatedly replicated. Silver layer is physically deposited on the PDMS. Then, gold nanoparticle (AuNP) inks are applied on the nanostructured PDMS using inkjet printer (Dimatix DMP 2831) to deposit AuNPs on the substrates. The characteristics of SERS-active substrates are measured; topology is provided by atomic force microscope (AFM, Park Systems XE-100) and Raman spectra are collected by Raman spectroscopy (Horiba LabRAM ARAMIS Spectrometer). We anticipate that the results may open up various possibilities of applying flexible platform to highly sensitive Raman detection.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.22 no.2
• /
• pp.372-379
• /
• 1998

A new 6 degrees-of-freedom micro stage, based on parallel mechanisms and actuated by using piezoelectric elements, has been developed for the application of micro positioning such as semiconductor manufacturing devices, high precision optical measurement systems, and high accurate machining. The micro stage structure consists of a base platform and an upper platform(stage). The base platform can effectively generates planar motion with yaw motion, while the stage can do vertical motion with roll and pitch motions with respect to the base platform. This separated structure has an advantage of less interference among actuators. The forward and inverse kinematics of the micro stage are discussed. Also, through linearization of kinematic equations about an operating point on the assumption that the configuration of the micro stage remains essentially constant throughout a workspace is performed. To maximize the workspace of the stage relative to fixed frame, an optimal design procedure of geometric parameter is shown. Hardware description and a prototype are presented. The prototype is about 150mm in height and its base platform is approximately 94mm in diameter. The workspace of the prototype is obtained by computer simulation. Kinematic calibration procedure of the micro stage and its results are presented.

• Transactions of Materials Processing
• /
• v.21 no.3
• /
• pp.179-185
• /
• 2012

In this paper, it is proposed to produce high precision screws with a diameter of $800{\mu}m$ and a thread pitch of $200{\mu}m$ ($M0.8{\times}P0.2$) by means of a cold thread rolling process. In this part II of the study, the focus is on the production and reliability testing of the prototype $M0.8{\times}P0.2$ micro-screw. Designs for two flat dies were developed with the aid of the literature and previous studies. Process parameters during the cold thread rolling process were established through FE simulations. The simulation results showed that the threads of the micro-screw are completely formed through the rolling process. Prototype $M0.8{\times}P0.2$ micro-screw were fabricated with a high precision thread rolling machine. In order to verify the simulation results, the deformed shape and dimensions obtained from the experiment were compared with those from the simulations. Hardness and failure torque of the fabricated micro-screw were also measured. The values obtained indicate that the CAE based process design used in this paper is very appropriate for the thread rolling of micro-sized screws.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.16 no.2
• /
• pp.14-22
• /
• 2007

In this paper, the acoustic emission(AE) behaviors were investigated with single-and 2-spot resistance spot welded SPCC specimens. Test specimens were welded horizontally and/or vertically according to the rolling direction of base netal in 2-spot welding. In the case of 2-spot welding, when tensile-shear test has below amplitudes: crack initiation $50{\sim}60dB;$ tear fracture $40{\sim}50dB$. And when cross tensile test has below amplitudes: early stage $75{\sim}85dB;$ yielding point $65{\sim}75dB;$ post yielding $40{\sim}60dB;$ plug fracture $70{\sim}80dB\;or\;90{\sim}100dB$. Also, from the b-value that is slope of AE amplitude, we knew that there are lots of low amplitudes if b-value is big(i.e., tensile-shear $specimen{\rightarrow}tear$ fracture or shear fracture), and there are lots of high amplitudes if b-value is small(i.e.. cross tensile $specimen{\rightarrow}plug$ fracture). As the results of fiacture mechanism analyses through AE amplitude distributions, change of the b-value represented fracture patterns of materials. Correspondingly, low amplitude signals appeared in crack initiation, and high amplitude signals appeared in base metal fracture. We confirmed that these amplitude distributions represented the change or degradation of materials.

• The Journal of Information Technology
• /
• v.3 no.4
• /
• pp.1-12
• /
• 2000

We can devide the speech into high frequency speech and low frequency speech according to the feature of the speech, However so far the construction of the recognizer without concerning this feature causes low recognition rate relatively and the needs of an amount of data in the research on the speech recognition. In this paper, we propose the method that can devide this feature of speaker's speech using the Formant frequency, and the method that can recognize the speech after constructing the recognizer model reflecting the feature of the high and low frequency of the speaker's speech, For the experiment we constructed the recognizer model using 47 mono-phone of Korean and trained the recognizer model using 20 women's and men's speech respectively. We divided the feature of speech using the Formant frequency Table, that had been consisted of the Formant frequency, and the value of pitch, and then We performed recognition using the trained model according to the feature of speech The proposed system outperformed the existing method in the recognition rate, as the result.

• The KIPS Transactions:PartA
• /
• v.11A no.4
• /
• pp.235-242
• /
• 2004

When manufacturing information and communication device that consists of lots of part, it is important to improve the quality of the produced system by inspecting the system accurately and exclude the defected part. In case of LCD which is recently in a great demand, the inspection process of the nut which bonds the back frame to protect the LCD panel has to be done by human labor. It highly required an automatic inspection system which can inspect the nut without wasting human resources. In this paper, we describe the process of developing a system which automatically inspect the status of nuts inserted during the manufacturing of LCD. The nut inspection vision system developed measures the number of nut's spiral, the distance between pitches, the width of a pitch, and the inside diameter of nut. We have adopted lens with high magnifying power and calibration tool and intended to produce automatic lighting for maintaining a stable environment for a high precision system. We also developed the algorithms for analyzing the nut. We apply the system to real factory field and verify that it is better than the man power in terms of error rate.

• Computers and Concrete
• /
• v.19 no.1
• /
• pp.59-68
• /
• 2017

This study focused on modeling the behavior of the compressive stress using the average strain and ultrasonic test results in concrete. Feed-forward backpropagation artificial neural network (ANN) models were used to compare four types of concrete mixtures with varying water cement ratio (WC), ordinary concrete (ORC) and concrete with short steel fiber-reinforcement (FRC). Sixteen (16) $150mm{\times}150mm{\times}150mm$ concrete cubes were used; each contained eighteen (18) data sets. Ultrasonic test with pitch-catch configuration was conducted at each loading state to record linear and nonlinear test response with multiple step loads. Statistical Spearman's rank correlation was used to reduce the input parameters. Different types of concrete produced similar top five input parameters that had high correlation to compressive stress: average strain (${\varepsilon}$), fundamental harmonic amplitude (A1), $2^{nd}$ harmonic amplitude (A2), $3^{rd}$ harmonic amplitude (A3), and peak to peak amplitude (PPA). Twenty-eight ANN models were trained, validated and tested. A model was chosen for each WC with the highest Pearson correlation coefficient (R) in testing, and the soundness of the behavior for the input parameters in relation to the compressive stress. The ANN model showed increasing WC produced delayed response to stress at initial stages, abruptly responding after 40%. This was due to the presence of more voids for high water cement ratio that activated Contact Acoustic Nonlinearity (CAN) at the latter stage of the loading path. FRC showed slow response to stress than ORC, indicating the resistance of short steel fiber that delayed stress increase against the loading path.