• Proceedings of the Korean Institute of Resources Recycling Conference
• /
• 2005.10a
• /
• pp.298-303
• /
• 2005

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.256-261
• /
• 2003

A memory industry is developing rapidly according to the period of the ubiquitous to approach. According to the development of a memory industry, the efficiency of the manufacture is becoming the serious consideration. DC parameter test system was a development low in this research for an efficiency increase of the manufacture. DC parameter test system increase of the manufacture. In the method to measure the output after permit volt and current at element.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10b
• /
• pp.1597-1600
• /
• 2001

In-Service-Monitoring을 통하여 네트웍 계층에서 성능저하와 전달되어지는 QOS의 영향을 감시하여 Pro-Active 매니지먼트가 가능하다. 그 결과로서 SLA 보증을 위반하기 전에 성능저하를 조기 발견하여 행동을 취한 수 있게 한다. 본 논문에서는 사용자의 요구조건을 만족하기 위한 네트웍의 구성과 SLA 측정의 제약조건들을 고려하고 Policy Based Routing이 적용되는 네트웍에서 Technology-Specific Parameter인 IP Packet Loss Ratio, IP 패킷전송지연, 패킷지터, 가용도등의 SLA Parameter를 측정하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.446-448
• /
• 2021

In training a deep learning model, it is crucial to tune various hyperparameters and gain speed and accuracy. While hyperparameters that mathematically induce convergence impact training speed, system parameters that affect host-to-device transfer are also crucial. Therefore, it is important to properly tune and select parameters that influence the data loader as a system parameter in overall time acceleration. We propose an automated framework called Num Worker Tuner (NWT) to address this problem. This method finds the appropriate number of multi-processing subprocesses through the search space and accelerates the learning through the number of subprocesses. Furthermore, this method allows memory efficiency and speed-up by tuning the system-dependent parameter, the number of multi-process spawns.

• Korea Information Processing Society Review
• /
• v.7 no.5
• /
• pp.36-40
• /
• 2000

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.36 no.8
• /
• pp.594-602
• /
• 1987

This paper proposes a new method of reducing the processing time and the size of consummimg memories in Hough transform. In this method, only the functional patterns are considered. The candidate points which are accumulated into the parameter space are computed in a many-to-one fashion and the parameter space is quantized dynamically to maintain a fine precision where it is needed. And a coarse-to-fine extraction method is used to reduce the processing time. The many-to-one fashional computation results in a relatively high-densed accumulation of candidate points around the parameter points corresponding to the image patterns in the image space. So, the dynamic quantization procedure can be simplified and the local maxima can be determined easily. And more effective reduction can be obtained as the dimension of parameter space is increased.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.512-513
• /
• 2023

CNN 기반 Deep Learning 분야에서 객체 탐지 정확도를 높이기 위해 모델의 많은 Parameter 가 사용된다. 많은 Parameter 를 사용하게 되면 최소 하드웨어 성능 요구치가 상승하고 처리속도도 감소한다는 문제가 있어, 최소한의 정확도 하락으로 Parameter 를 줄이기 위한 여러 Pruning 기법이 사용된다. 본 연구에서는 Neural Architecture Search(NAS) 기반 Channel Pruning 인 Artificial Bee Colony(ABC) 알고리즘을 사용하였고, 기존 NAS 기반 Channel Pruning 논문들이 Classification Task 에서만 실험한 것과 달리 Object Detection Task 에서도 NAS 기반 Channel Pruning 을 적용하여 기존 Uniform Pruning 과 비교할 때 파라미터 수 대비 정확도가 개선됨을 확인하였다.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.401-404
• /
• 2004

SAR (Synthetic Aperture Radar) signal data need a focusing procedure to make the information available to the user. In recent SAR systems, various sensing modes and mission operations are applied to acquire high-resolution SAR images. Therefore, in order to develop generalized focusing software for multi-satellites, a regularized parameter configuration that sufficiently represents sensor and platform characteristics of the SAR system is required. The objective of this paper is to introduce the consideration of parameter definition for developing a generalized SAR processor and to discuss the flexibility and extensibility of defined parameters. The proposed parameter configuration can be applied to a SAR processor. Experiments based on real data will show the suitability of the suggested processing parameters.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.1150-1155
• /
• 2008

Parameter analysis has been performed for an equal channel angular process. The processing variables such as channel configuration, friction coefficient, and ram speed were investigated by means of the magnitudes and distributions of effective plastic strain analysis through the deformation. The materials considered were pure aluminum and titanium. Here firstly, a finite element implementation by using the commercial ABAQUS software was carried out for both the aluminum and titanium materials based on the L-channel configuration. The experimental investigation then has been conducted using the obtained data. Finally, the ability of robust metals which can be produced by the optimized ECAP has been discussed by the appropriate parameter analysis.

• Laser Solutions
• /
• v.10 no.1
• /
• pp.11-17
• /
• 2007

We presented the proper laser processing conditions for the capillary line marking, which could be applied for the fabrication of injection needles. With changing the parameters such as lamp current, duty cycle and beam amplification factor of beam expander, we evaluated the processing performance considering amount of dross, processing efficiency and processed linewidth in the sample. We could carry out the proper line marking at the condition of 70% lamp current, duty cycle of 7-10% and 6-times amplification of beam diameter. To perform efficient line processing, the utilization of duty cycle of 12% at 80% lamp current was also preferred.