• Journal of Multimedia Information System
• /
• v.3 no.3
• /
• pp.97-102
• /
• 2016

Deep learning enables machines to have perception and can potentially outperform humans in the medical field. It can save a lot of time and reduce human error by detecting certain patterns from medical images without being trained. The main goal of this paper is to build the optimal model for breast mass classification by applying various methods that influence the performance of Convolutional Neural Network (CNN). Google's newly developed software library Tensorflow was used to build CNN and the mammogram dataset used in this study was obtained from 340 breast cancer cases. The best classification performance we achieved was an accuracy of 0.887, sensitivity of 0.903, and specificity of 0.869 for normal tissue versus malignant mass classification with augmented data, more convolutional filters, and ADAM optimizer. A limitation of this method, however, was that it only considered malignant masses which are relatively easier to classify than benign masses. Therefore, further studies are required in order to properly classify any given data for medical uses.

• Proceedings of the KIEE Conference
• /
• 2006.04b
• /
• pp.95-98
• /
• 2006

This paper proposes high speed SRM drive system for blower with a new 4-level inverter and precise excitation position generator. For the high speed blower, a proper inverter and control method are proposed and the output characteristics are analyzed. In order to get a fast build-up and demagnetization of excitation a current, 4-level inverter system is proposed. The proposed 4-level inverter has additional charge capacitor, power switch and diode in the conventional asymmetric converter. The charged high voltage is supplied to the phase winding for fast current build-up, and demagnetization current is charged to additional capacitor of 4-level inverter. In addition, a precise excitation position generator can reduce turn-on and turn-off angle error according to sampling period of digital control system. The proposed high speed SRM drive system is verified by computer simulation and experimental result.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.55 no.7
• /
• pp.337-345
• /
• 2006

This paper proposes a high speed SRM drive system for blower application with a new 4-level inverter and precise excitation position generator. For a high speed blower, a proper inverter and control method are proposed and the output characteristics are analyzed. In order to get a fast build-up and demagnetization of excitation current, a 4-level inverter system is proposed. The proposed 4-level inverter has additional charge capacitor, power switch and diode in the conventional asymmetric converter. The charged high voltage is supplied to the phase winding for fast current build-up, and demagnetization current is charged to additional capacitor of the 4-level inverter. In addition, a precise excitation position generator can reduce turn-on and turn-off angle error according to sampling period of digital control system. The proposed high speed SRM drive system is verified by computer simulation and experimental results.

• The Journal of Korean Society for Radiation Therapy
• /
• v.6 no.1
• /
• pp.89-93
• /
• 1994

Authors have measured virtual source distance of electron beam from CL/1800 medical linear accelerator, with newly designed method. Beam scanning was performed with the direction of beam axis in the air. Compared results between this study and well established in phantom measurement shows good agreement with in experimental error. And we have found that build-up cap plays very important role in air measurement because of charge build up. The method of in-air measurement of virtual source distance is very easy to set-up and generate accurate results.

• Proceedings of the KIEE Conference
• /
• 2005.10c
• /
• pp.110-113
• /
• 2005

This paper proposes high speed SRM drive system for blower with a new 4-level inverter and precise excitation position generator. For the high speed blower, a proper 12/8 SRM is designed and analyzed. In order to get a fast build-up and demagnetization of excitation a current, now 4-level inverter system is proposed. The proposed 4-level inverter has additional charge capacitor, power switch and diode in the conventional asymmetric converter. The charged high voltage is supplied to the phase winding for fast current build-up, and demagnetization current is charged to additional capacitor of 4-level inverter. In addition, a precise excitation position generator can reduce turn-on and turn-off angle error according to sampling period of digital control system. The proposed high speed SRM drive system is verified by computer simulation.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.219-222
• /
• 2003

In this paper, we study the usage of phoneme duration information for rejection garbage sentence. First, we build a phoneme duration modeling in a speech recognition system based on dicicion tree state tying, We assume that phone duration has a Gamma distribution. Next, we build a verification module in which word-level confidence measure is used. Finally, we make a comparative study on phoneme duration with speech DB obtained from the live system. This DB consistes of OOT(out-of-task) and ING(in-grammar) utterences. the usage of phone duration information yields that OOT recognition rate is improved by 46％ and that another 8.4％ error rate is reduced when combined with utterence verification module.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.11
• /
• pp.1626-1634
• /
• 1993

In this paper we propose a coding scheme which combines source codes and error correcting codes in order to be robus to channel noise. One of the coding schemes that take into account both the source and the channel is the channel optimized quantizer (COQ) which simultaneously minimizes quantization noise and the noise due to channel errors. This paper deals with the problem of combining channel optimized quantizers with ECC to build an improved system. To be specific, we computed the performance of an n bit COQ and that of an n-1 bit COQ followed by an (n-1)/n punctured convolutional code. From this result whether or not the ECC are selected is determined by the number of allocated bits and the channel bit error rate. These results are applied to the image trans-mission method using DCT, and the system performances are evaluated.

• Journal of the Korean Society for Precision Engineering
• /
• v.19 no.3
• /
• pp.152-159
• /
• 2002

SLA (Stereolithography Apparatus) it a process used to rapidly produce polymer components directly from a computer representation of the part. Though SLA is being recognized as an innovative technology, it still cannot be used to fully practical application since it lacks of dimensional accuracy compared to conventional process. If the shrinkage were perfectly uniform and no distortion took place, excellent part accuracy could still be achieved through and appropriate scaling factor when generating the build file. However, in certain geometries involving intersecting thick and thin sections, nonuniform retrain shrinkage becomes the engine of part distortion. In order to improve the part accuracy of SLA, this paper evaluates how largely each parameter of SLA contributes to the part accuracy and estimates the optimal set of parameter which minimizes the dimension error of the test part, "Slab (100mm$\times$100mm$\times$2mm)"and "scale bar"part. Three control parameters such as critical exposure, generation depth and fill cure depth are used.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.10a
• /
• pp.347-349
• /
• 2014

Model verification and validation (V&V) is a current research topic to build computational models with high predictive capability by addressing the general concepts, processes and statistical techniques. The hypothesis test for validity check is one of the model validation techniques and gives a guideline to evaluate the validity of a computational model when limited experimental data only exist due to restricted test resources (e.g., time and budget). The hypothesis test for validity check mainly employ Type I error, the risk of rejecting the valid computational model, for the validity evaluation since quantification of Type II error is not feasible for model validation. However, Type II error, the risk of accepting invalid computational model, should be importantly considered for an engineered products having high risk on predicted results. This paper proposes a technique named as the response-adaptive experimental design to reduce Type II error by adaptively designing experimental conditions for the validation experiment. A tire tread block problem and a numerical example are employed to show the effectiveness of the response-adaptive experimental design for the validity evaluation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.6
• /
• pp.2575-2582
• /
• 2013

In the delivery systems for public construction works, design-build methods such as turn-key and alternative have contributed to improving the quality of structures and developing the construction technology. However, it is pointed out that construction based on a design-build method is likely to waste a budget due to its higher contract price than construction based on a design-bid-build method. Other issues such as contract monopoly, bid rigging and deliberation irregularity in design-build. Since there are not much works have been accomplished to comparatively analyze the actual performance and effect based on delivery methods, discussing and judging the performance and effect of each deliver method with limited information could lead to an error. Thus, this study by extensively investigates the actual project cost data, ranging from contract price to the construction cost of the public construction works and comparatively analyzed the characteristics of each delivery method about cost structure and fluctuation trend. This work is expected to assist stake-holders in properly understanding delivery systems for public construction works and promoting their efficient management.