• 전자공학회논문지SC
• /
• 제47권5호
• /
• pp.24-32
• /
• 2010

본 연구에서는 Optical Flow Method의 추적 성능을 향상시키기 위한 전처리 알고리즘을 제안한다. 제안하는 알고리즘은 Median Filter, Binarization, Morphology, Canny Edge, Contour Detecting 및 Approximation Method를 기반으로 개발되었다. Optical Flow 추적 성능 향상 여부를 평가하기 위해 Lucas-Kanade Optical Flow 알고리즘에 개발된 전처리 알고리즘을 적용하고, 전처리 알고리즘이 적용되지 않은 Optical Flow 영상과 추적 결과를 비교 분석하였다. 또한, Median Filter와 Histogram Equalization으로 구성된 기존 전처리 알고리즘과의 결과 비교를 통해, 개발된 전처리 알고리즘의 추적 성능 향상여부를 평가하였다. 실험결과, 전처리 알고리즘을 적용하지 않은 영상과 기존 전처리 알고리즘을 적용한 영상은 특징영역의 분할이 이루어지지 않아, Optical Flow의 추적 정확도가 매우 낮게 나타났다. 반면, 개발된 전처리 알고리즘을 적용한 영상에서는 외곽선이 내외부로 세분화되고, 외곽선 트리가 구성됨에 따라 Optical Flow의 추적 성능이 매우 높게 나타났다.

• 한국시뮬레이션학회논문지
• /
• 제21권2호
• /
• pp.59-67
• /
• 2012

그리드 컴퓨팅은 하나의 대용량 작업을 처리하도록 물리 자원을 구성하고 있지만 최근에는 데이터의 급속한 증가로 인해서 복수개의 작업을 처리하는 방법이 필요하다. 일반적으로 대용량 작업을 요청하면 각 물리 자원들이 작업을 분할하게 되며, 자원의 성능과 거리에 따라 처리 시간이 다르다. 성능에 따라 먼저 완료된 유효자원은 어떠한 작업도 하지 않으며, 모든 작업이 끝났을 경우에 다음 작업을 처리한다. 이에 본 논문에서는 먼저 처리가 완료된 자원을 다른 작업에 할당할 수 있는 동적 자원 재배치 스케줄링 모델(DRRSM: Dynamic Resource Reallocation Scheduling Model)을 제안한다. DRRSM은 먼저 처리가 완료된 자원을 다른 작업에 자원의 성능과 거리에 따라 작업을 재배치시키는 방법이다. DRRSM은 여러 개의 대용량 작업을 처리하는데 효과적이다.

• Architectural research
• /
• 제15권3호
• /
• pp.167-174
• /
• 2013

Due to increasing competition of construction companies, it is required to optimize the management of construction projects and "lean" concepts are rapidly spreaded in construction industry. Steel work accounts for a large proportion in construction work, and a variety of attempts to efficiently perform steel work has currently made. And since rebar (Engineer-to-Order) can be engineered through design once order is placed, it gives the great impact on construction, thus, entire management is required. The purpose of this study is to present the method to increase the efficiency of field processing method of steel work in terms of lean construction. Once we examine process of steel work and identify the flow, we would like to analyze which processes should be improved through value analysis approach and present the improvement plans. Also, this study examines cases of field processing and factory processing, and it identifies the waste factors in the procurement process. Finally, this study would like to present the result of analysis from the perspectives of value. The rebar delivery process is divided into several steps and the duration of every step is surveyed. Using duration data, VAT (Value added time) can be calculated for analyzing the efficiency of the process.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2017년도 춘계학술대회
• /
• pp.645-648
• /
• 2017

기존의 Embedded Tester는 Log 처리를 위해 TCP와 공유 파일 시스템을 이용한 Server - Client간 1-N 구조로 처리 되었다. 이러한 방식은 구현 난이도에 따른 시간적 손실과 예외처리에 따른 Tester의 리소스 낭비가 발생한다. 이에 메시징 시스템을 이용하여 분산처리가 가능한 Log 처리 메시지 레이어를 구현하고 기존의 TCP, 공유 파일 시스템 전송방식과 비교하였다. 비교 결과 메시지 레이어를 이용한 전송이 TCP 보다 더 높은 전송 대역폭을 보였다. CPU 사용량에서 메시지 레이어가 TCP 보다 낮은 효율을 보였으나 큰 차이를 보이지 않았다. 이를 통해 메시지 레이어를 이용한 Log 처리가 더 높은 효율을 보임을 알 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제3권4호
• /
• pp.49-54
• /
• 1998

능동규칙의 조건부 처리에 사용되는 데이터베이스 연산들은 처리되는 시간들이 상당히 크기 때문에 능동 데이터베이스 시스템의 성능에 중요한 영향을 미친다. 그래서 조건부에서 발생되는 연산들의 처리 시간을 최소화 시켜야만 시스템의 성능을 높일 수 있다. 기존 연구의 처리 기법들은 기본 데이터베이스 연산과 한정된 일부 집계함수만을 효율적으로 처리하고 있다. 본 연구에서는 릴레이션의 구조화와 상태테이블을 사용한 처리 기법을 제시한다. 릴레이션들을 분류트리로 구조화시키고, 삭제 정보 테이블을 도입함으로써 기본 데이터베이스 연산의 처리시간을 줄일 수 있다. 또한 이진 검출트리와 릴레이션의 상태테이블을도입함으로써 처리비용이 큰 집계함수를 효율적으로 처리할 수 있어 능동 데이더베이스 시스템의 이용을 극대화시킬 수 있다.

• 품질경영학회지
• /
• 제48권4호
• /
• pp.553-566
• /
• 2020

Purpose: The purpose of this study was to design a framework for generating one-class classification algorithm based on Hyper-Rectangle(H-RTGL) in a distributed environment connected by network. Methods: At first, we devised one-class classifier based on H-RTGL which can be performed by distributed computing nodes considering model and data parallelism. Then, we also designed facilitating components for execution of distributed processing. In the end, we validate both effectiveness and efficiency of the classifier obtained from the proposed framework by a numerical experiment using data set obtained from UCI machine learning repository. Results: We designed distributed processing framework capable of one-class classification based on H-RTGL in distributed environment consisting of physically separated computing nodes. It includes components for implementation of model and data parallelism, which enables distributed generation of classifier. From a numerical experiment, we could observe that there was no significant change of classification performance assessed by statistical test and elapsed time was reduced due to application of distributed processing in dataset with considerable size. Conclusion: Based on such result, we can conclude that application of distributed processing for generating classifier can preserve classification performance and it can improve the efficiency of classification algorithms. In addition, we suggested an idea for future research directions of this paper as well as limitation of our work.

• 대한임베디드공학회논문지
• /
• 제18권4호
• /
• pp.165-172
• /
• 2023

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

• 전기학회논문지P
• /
• 제67권3호
• /
• pp.137-142
• /
• 2018

Energy harvesting technology is drawing attention as a means of collecting various eco-friendly energy and accumulating residual energy. Recently, differential power processing (DPP) is being developed as part of energy harvesting. This is being studied as a solution to the loss of power generation between power modules and the problems caused by module small losses depending on the size of power production. In this paper, we propose the necessity of the DPP by comparing and analyzing energy harvesting related module integration system and power supply efficiency of DPP. The power efficiency of the converter and the power difference between the wind power and the photovoltaic power supply have been changed to demonstrate the effectiveness of the proposed system.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1149-1157
• /
• 2017

The satisfiability problem is always a core problem in artificial intelligence (AI). And how to improve the efficiency of algorithms solving the satisfiability problem is widely concerned. Algorithm IER (Improved Extension Rule) is based on extension rule. The number of atoms and the number of clauses affect the efficiency of the algorithm IER. DPLL rules are helpful to reduce these numbers. Then a complete algorithm CIER based on splitting rule and extension rule is proposed in this paper in order to improve the efficiency. At first, the algorithm CIER (Complete Improved Extension Rule) reduces the scale of a clause set with DPLL rules. Then, the clause set is split into a group of small clause sets. In the end, the satisfiability of the clause set is got from these small clause sets'. A strategy MOAMD (maximum occurrences and maximum difference) for the algorithm CIER is given. With this strategy, a better arrangement of atoms could be got. This arrangement could make the number of small clause sets fewer and the scale of these sets smaller. So, the algorithm CIER will be more efficient.

• Journal of Information Processing Systems
• /
• 제15권2호
• /
• pp.271-287
• /
• 2019

The synchronization scheme based on moving average is robust and suitable for the same rule to be adopted in embedding watermark and synchronization code, but the imperceptibility and search efficiency is seldom reported. The study aims to improve the original scheme for robust audio watermarking. Firstly, the survival of the algorithm from desynchronization attacks is improved. Secondly, the scheme is improved in inaudibility. Objective difference grade (ODG) of the marked audio is significantly changed. Thirdly, the imperceptibility of the scheme is analyzed and the derived result is close to experimental result. Fourthly, the selection of parameters is optimized based on experimental data. Fifthly, the search efficiency of the scheme is compared with those of other synchronization code schemes. The experimental results show that the proposed watermarking scheme allows the high audio quality and is robust to common attacks such as additive white Gaussian noise, requantization, resampling, low-pass filtering, random cropping, MP3 compression, jitter attack, and time scale modification. Moreover, the algorithm has the high search efficiency and low false alarm rate.