• 한국항해항만학회지
• /
• 제33권8호
• /
• pp.555-562
• /
• 2009

컨테이너 터미널의 주요 생산성 지표는 안벽에서의 작업 효율성이라 할 수 있다. 안벽에서는 Q/C(Quay Crane)이라는 장비가 접안 선박의 컨테이너를 하역한다. Q/C의 작업 생산성을 높이기 위해서는 좀 더 효율적인 Y/T(Yard Tractor)운영 방식이 필요하다. 기존 작업 방식(싱글 사이클)에서는 양하작업 이후 적하 작업이 이루어진다. 듀얼 사이클이란 양하작업과 적하 작업을 동시에 함으로써 안벽 생산성과 야드 트랙터의 이용률을 높이는 방법이다. 터미널에서 듀얼 사이클의 도입은 추가적인 장비의 도입 없이 운영에서의 변화만을 요구한다. 즉, 기존의 dedicate 시스템에서 pooling 시스템으로의 변화가 필요하다. 본 논문에서는 듀얼 사이클을 이용하는 항만에서의 작업 효율성을 증대시키기 위한 선적 계획 방법을 제시하고자 한다. 이 문제를 풀기위해 유전 알고리즘과 타부서치를 제시하였다.

• International Journal of Computer Science & Network Security
• /
• 제22권5호
• /
• pp.175-181
• /
• 2022

COVID-19 is an acute respiratory syndrome that affects the host's breathing and respiratory system. The novel disease's first case was reported in 2019 and has created a state of emergency in the whole world and declared a global pandemic within months after the first case. The disease created elements of socioeconomic crisis globally. The emergency has made it imperative for professionals to take the necessary measures to make early diagnoses of the disease. The conventional diagnosis for COVID-19 is through Polymerase Chain Reaction (PCR) testing. However, in a lot of rural societies, these tests are not available or take a lot of time to provide results. Hence, we propose a COVID-19 classification system by means of machine learning and transfer learning models. The proposed approach identifies individuals with COVID-19 and distinguishes them from those who are healthy with the help of Deep Visual Embeddings (DVE). Five state-of-the-art models: VGG-19, ResNet50, Inceptionv3, MobileNetv3, and EfficientNetB7, were used in this study along with five different pooling schemes to perform deep feature extraction. In addition, the features are normalized using standard scaling, and 4-fold cross-validation is used to validate the performance over multiple versions of the validation data. The best results of 88.86% UAR, 88.27% Specificity, 89.44% Sensitivity, 88.62% Accuracy, 89.06% Precision, and 87.52% F1-score were obtained using ResNet-50 with Average Pooling and Logistic regression with class weight as the classifier.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 추계학술발표대회
• /
• pp.30-33
• /
• 2020

최근 엣지 컴퓨팅과 같은 임베디드 디바이스에서 CNN과 같은 딥러닝 모듈을 수행하기 위해서 하드웨어 설계 및 구현이 많이 진행되고 있다. 이러한 임베디드 시스템에 필요한 CNN모듈을 위한 하드웨어 설계를 위해서 먼저 모델링을 통해서 시뮬레이션이 필요하다. 본 논문에서는 오픈 라이센스를 이용한 RISC-V로 딥러닝 시뮬레이터를 제작하였다. SystemC로 구현된 RISC-V를 Virtual Platform로 시뮬레이터의 제작을 하여 시뮬레이팅을 하였고, SystemC의 특징인 모듈화와 모듈간 통신에 유의하여 시스템을 구성하였다. CNN 알고리즘을 참조하여 Convolution, Activation, Pooling 연산의 기능을 하는 시스템을 구성하였다.

• 한국정보처리학회논문지
• /
• 제7권3호
• /
• pp.909-922
• /
• 2000

High sped messaging layer for application's feeling of low level network performance is needed by Clustering System based on high speed network fabrics. It should have the mechanism to directly pass messages between network card and application space, and provide flexible affodabilities for many diverse applications. In this paper, CROWN (Clustering Resources On Workstations' Network) which is designed and implemented for multi-purpose clustering system will be introduced briefly, and CLCP(CROWN Lean Communication Primitives)which is the high speed messaging layer for CROWN will be followed. CLCP consists of a firmware for controlling Myrinet card, device drier, and user libraries. CLCP supports various application domains as a result of pooling and interrupt receive mechanism. In case of polling based receive, 8 bytes short message, and no other process, CLCP has 262 micro-second response time between two nodes, and IM bytes large message, it shows 442Mbps bandwidth.

• 한국항해항만학회지
• /
• 제36권10호
• /
• pp.895-904
• /
• 2012

항만간 허브항 경쟁이 극심해 지고 있는 오늘날, 컨테이너 선사는 M&A 및 전략적 제휴로 컨테이너터미널 운영사와의 가격 협상력의 우월적 지위를 갖게 되어 컨테이너터미널 운영사간에 선사 및 화물유치를 위한 경쟁을 더욱 부추기고 있다. 그러나 수요측면에서 컨테이너물동량 증가율 둔화로 컨테이너터미널에서 처리해야 할 물동량은 한정되어 있는 반면, 공급 측면에서 항만터미널의 지속적인 건설은 항만간 또는 터미널간 물량 유치경쟁을 과열시키고 있다. 특히 부산항은 신항 개장이후 북항과 신항간 물동량 유치경쟁으로 인하여 항만하역시장의 교란을 가져오고 있다. 본 연구에서는 부산항 컨테이너 항만하역시장의 구조적 특성분석과 설문조사 방법론을 통하여 향후 부산항 항만 하역시장의 안정화 방안을 제시하고자 한다. 하역시장 안정화 방안으로 항만하역 요율결정체계 개선, 요금신고제도의 개선 및 공정경쟁규약의 제정 등과 같은 법 제도적 개선방안과 컨테이너터미널 운영사별 처리물량 상한제를 도입을 전제로한 항만풀링제도 및 물량연동 임대료제도의 도입방안을 제안한다.

• BMB Reports
• /
• 제39권4호
• /
• pp.418-425
• /
• 2006

A human bacterial artificial chromosome (BAC) library was constructed with high molecular weight DNA extracted from the blood of a male Korean. This Korean BAC library contains 100,224 clones of insert size ranging from 70 to 150 kb, with an average size of 86 kb, corresponding to a 2.9-fold redundancy of the genome. The average insert size was determined from 288 randomly selected BAC clones that were well distributed among all the chromosomes. We developed a pooling system and three-step PCR screen for the Korean BAC library to isolate desired BAC clones, and we confirmed its utility using primer pairs designed for one of the clones. The Korean BAC library and screening pools will allow PCR-based screening of the Korean genome for any gene of interest. We also determined the allele types of HLA-DRA and HLA-DRB3 of clone KB55453, located in the HLA class II region on chromosome 6p21.3. The HLA-DRA and DRB3 genes in this clone were identified as the DRA*010202 and DRB3*01010201 types, respectively. The haplotype found in this library will provide useful information in future human disease studies.

• 한국인터넷방송통신학회논문지
• /
• 제21권5호
• /
• pp.141-147
• /
• 2021

본 연구를 통해 임베디드 시스템(Embedded System)에서 뉴럴 네트워크(Neural Network) 인풋의 차원 감소 방식으로 복잡한 연산량을 줄여 초분광 대용량 데이터 특징 정보의 활용률을 개선하기 위해, 전체 밴드를 밴드별 최댓값과 최솟값 차이로 부분집합으로 군집화하여, 각 부분집합에서 밴드 선택 알고리즘을 적용한다. 특징 추출과 특징 선택 기법 중에, 특징 선택 기법을 통해, 파장 범위와 관계없이 데이터세트에 맞는 최적의 밴드 수와 기존 알고리즘 적용 소요 시간과 성능을 향상하고자 한다. 이 실험을 통해 기존 밴드 선택 기법보다 1/3~ 1/9배 소요 시간을 단축했음에도 불구하고 K-최근접 이웃 분류기를 통한 성능 면에서는 약 4% 이상 향상된 의미 있는 결과를 도출하였다. 실시간 초분광 데이터 분석 활용에는 어렵지만, 개선된 가능성을 확인했다.

• Management Science and Financial Engineering
• /
• 제6권2호
• /
• pp.47-67
• /
• 2000

It is discussed how to assign delivery tasks to automated guided vehicles (AGVs) for multiple container cranes in automated container terminals. The primary goal of dispatching AGVs is to complete all the lading and discharging operations as early as possible, and the secondary goal is to minimize the total travel distance of AGVs. It is assumed that AGVs are not dedicated to a specific container crane but shared among multiple cranes. A mathematical formulation is developed and a heuristic algorithm is suggested to obtain a near optimal solution with a reasonable amount of computational time. The single-cycle and the dual-cycle operations in both the seaside and the landside operations are analyzed. The effects of pooling AGVs for multiple container cranes on the performance of an entire AGV system are also analyze through a numerical experiment.

• International Journal of Advanced Culture Technology
• /
• 제5권1호
• /
• pp.83-89
• /
• 2017

In this paper, we propose an extended method of one-hot Long Short-Term Memory (LSTM) and evaluate the performance on spam filtering task. Most of traditional methods proposed for spam filtering task use word occurrences to represent spam or non-spam messages and all syntactic and semantic information are ignored. Major issue appears when both spam and non-spam messages share many common words and noise words. Therefore, it becomes challenging to the system to filter correct labels between spam and non-spam. Unlike previous studies on information filtering task, instead of using only word occurrence and word context as in probabilistic models, we apply a neural network-based approach to train the system filter for a better performance. In addition to one-hot representation, using term weight with attention mechanism allows classifier to focus on potential words which most likely appear in spam and non-spam collection. As a result, we obtained some improvement over the performances of the previous methods. We find out using region embedding and pooling features on the top of LSTM along with attention mechanism allows system to explore a better document representation for filtering task in general.

• 수산해양교육연구
• /
• 제25권1호
• /
• pp.40-52
• /
• 2013

The purpose of this study is to explore improving ways of screening of self-directed learning in self-governing high schools for expanding self-initiated learning screening system. To achieve the above aim, it was used survey method to 99 admission teachers of 12 self-governing high schools and analyzed responses with SPSS program. The research findings are as follows. To expand self-initiated learning screening system in self-governing high schools, it needs to obtain admission members of entrance committee, pooling of external screening members, quick decision of school admission members, increasing payment, training and understanding of the system and consideration of admission documents and study plan. Some improving ways of the expanding of self-initiated learning screening may be listed as follows: 1) development of screening model for self-governing schools 2) policy consideration for self-governing public high school 3) expanding autonomy of school 4) supporting budget 5) sharing of operation case and reinforcement of public relations.