• 한국통신학회논문지
• /
• 제29권4C호
• /
• pp.533-540
• /
• 2004

본 논문에서는 웨이블릿(wavelet) 영역에서 분류 예측, KLT (Karhunen-Loeve transform), 및 3-D SPIHT(three-dimensional set partitioning in hierarchical trees) 알고리즘(algorithm)을 이용하여 인공위성 화상 데이터에 존재하는 대역내 중복성 (intraband redundancy)과 대역간 중복성 (interband redundancy)을 효과적으로 제거하는 새로운 압축 방법을 제안하였다. 대역간 중복성을 제거하기 위해 웨이블린 영역에서의 분류 정보를 이용하여 영역별 대역간 예측을 행한다. 영역별 대역간 예측에 의해 복원되는 화상들은 예측 오차로 인해 원 화상 (original image)과 차 화상 (residual image)을 가진다. 이 차 화상들 간에 존재하는 대역간 중복성을 제거하기 위하여 KLT를 행한다. 웨이블릿 변환 (wavelet transform)과 KLT를 행하여 대역내 및 대역간 크기 순서로 재정렬된 변환 계수들을 3-D SPIHT 알고리즘을 이용하여 부호화 한다. 제안한 방법의 성능 평가를 위해서 다분광 화상 데이터에 대하여 압축 실험을 행하여 제안한 방법이 기존의 방법들 보다 동일한 여러 비트율 (bit rate)에서 평균 PSNR (peak signal-to-noise ratio)이 0.12∼3.83㏈ 향상됨을 확인하였다.

• 한국통신학회논문지
• /
• 제25권7A호
• /
• pp.1028-1036
• /
• 2000

본 논문에서는 웨이블릿 변환을 이용한 오류에 강인한 영상 압축 기법을 제안하였다. 공간·주파수 영역에서 웨이블릿 계수들의 통계적 특성, 에너지 특성, 방향 특성을 이용한 제로트리 기법은 높은 압축 성능을 나타낸다. 하지만, 제로트리 부호와 기법은 부호에 따른 부호화 되는 계수의 수가 다르기 때문에 오류에 민감하게 반응하여 한 개의 오류가 전체 영상에 확산되어 영향을 미치게 된다. 제안 알고리듬에서는 SPIHT(Set Partitioning in Hierachical Trees) 알고리듬을 이용한 제로트리 기법으로 영상을 부호화한다. 그리고 부호화 계수들을 부밴드간 상관도를 이용하여 비트열을 다수의 블록으로 분리하고 비트 재구성 알고리듬을 이용하여 같은 크기의 블록으로 만든다. 이 과정으로 효율적인 비트 할당과 오류의 전파를 해당 블록으로 제한하여 오류가 없거나 적은 환경에서는 제로트리 압축 기법과 유사한 성능을 보이며 오류가 많은 환경에서는 제로트리 압축 기법 및 기존의 오류에 강인한 압축보다 더 효율적으로 부호화 할 수 있다.

• 정보과학회 논문지
• /
• 제43권1호
• /
• pp.40-46
• /
• 2016

도로지도 생성은 인공위성 촬영이나 현장실사를 기반으로 한다. 그리하여 도로지도를 생성하고 수정하는데 많은 시간과 비용이 든다. 이러한 이유로 차량 GPS 데이터를 이용해 도로지도를 생성하는 연구가 활발히 진행되고 있다. 도로지도 생성 연구에서 가장 중요한 문제는 주도로와 같은 대표궤적을 추출하는 것이다. 대표궤적 추출을 수행할 때에는 시작과 끝이 비슷한 궤적데이터들의 집합을 전제로 하여 궤적을 추출한다. 따라서 대표궤적을 추출하기에 앞서 전처리 과정으로 궤적 클러스터링 작업이 필요하다. 본 논문에서는 이러한 문제를 해결하기 위해 하나의 영역을 일정한 격자로 분할하고, Sweep Line 알고리즘을 응용해 유사궤적들을 탐색한다. 마지막으로 프레쉐거리를 이용하여 궤적 간 유사도를 계산하였다. 실제로 서울의 강남구 지역에 있는 500대의 차량 GPS 궤적을 가지고 클러스터링 작업을 수행하였다. 또한, 실험을 통하여 격자분할 접근방식의 빠른 수행시간과 안정성을 보였다.

• 경영과정보연구
• /
• 제9권
• /
• pp.25-39
• /
• 2002

We consider the Fuzzy clustering which is devised for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. The researches carried out in this field before show that the Fuzzy clustering concept is involved so much that for a certain set of data, the main purpose of the clustering cannot be attained as desired. Thus we propose a new objective function, named as Fuzzy-Entroppy Function in order to satisfy the main motivation of the clustering which is classifying the data clearly. Also we suggest Mean Field Annealing Algorithm as an optimization algorithm rather than the. ISODATA used traditionally in this field since the objective function is changed. We show the Mean Field Annealing Algorithm works pretty well not only for the new objective function but also for the classical Fuzzy objective function by indicating that the local minimum problem resulted from the ISODATA can be improved.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2000년도 춘계공동학술대회 논문집
• /
• pp.613-616
• /
• 2000

This paper considers the problem of subway crew scheduling. Crew scheduling is concerned with finding a minimum number of assignments of crews to a given timetable satisfying various restrictions. Traditionally, crew scheduling problem has been formulated as a set covering or set partitioning problem possessing exponentially many variables, but even the LP relaxation of the problem is hard to solve due to the exponential number of variables. In this paper, we propose two basic techniques that solve the problem in a reasonable time, though the optimality of the solution is not guaranteed. To reduce the number of variables, we adopt column-generation technique. We could develop an algorithm that solves column-generation problem in polynomial time. In addition, the integrality of the solution is accomplished by variable-fixing technique. Computational results show column-generation makes the problem of treatable size, and variable fixing enables us to solve LP relaxation in shorter time without a considerable increase in the optimal value. Finally, we were able to obtain an integer optimal solution of a real instance within a reasonable time.

• 한국경영과학회지
• /
• 제27권4호
• /
• pp.67-86
• /
• 2002

This paper considers subway crew scheduling problem. Crew scheduling is concerned with finding a minimum number of assignments of crews to a given timetable satisfying various restrictions. Traditionally, crew scheduling problem has been formulated as a set covering or set partitioning problem possessing exponentially many variables, but even the LP relaxation of the problem is hard to solve due to the exponential number of variables. In this paper. we propose two basic techniques that solve the subway crew scheduling problem in a reasonable time, though the optimality of the solution is not guaranteed. We develop an algorithm that solves the column-generation problem in polynomial time. In addition, the integrality of the solution is accomplished by variable-fixing technique. Computational result for a real instance is reported.

• 산업경영시스템학회지
• /
• 제25권6호
• /
• pp.36-41
• /
• 2002

We consider the Fuzzy clustering which is devised for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. The researches carried out in this field before show that the Fuzzy clustering concept is involved so much that for a certain set of data, the main purpose of the clustering cannot be attained as desired. Thus we propose a new objective function, named as Fuzzy-Entroppy Function in order to satisfy the main motivation of the clustering which is classifying the data clearly. Also we suggest Mean Field Annealing Algorithm as an optimization algorithm rather than the ISODATA used traditionally in this field since the objective function is changed. we show the Mean Field Annealing Algorithm works pretty well not only for the new objective function but also for the classical Fuzzy objective function by indicating that the local minimum problem resulted from the ISODATA can be improved.

• 경영과정보연구
• /
• 제14권
• /
• pp.51-65
• /
• 2004

We consider the Fuzzy clustering which is devised for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. The researches carried out in this field before show that the Fuzzy clustering concept is involved so much that for a certain set of data, the main purpose of the clustering cannot be attained as desired. Thus we Propose a new objective function, named as Fuzzy-Entroppy Function in order to satisfy the main motivation of the clustering which is classifying the data clearly. Also we suggest Mean Field Annealing Algorithm as an optimization algorithm rather than the ISODATA used traditionally in this field since the objective function is changed. We show the Mean Field Annealing Algorithm works pretty well not only for the new objective function but also for the classical Fuzzy objective function by indicating that the local minimum problem resulted from the ISODATA can be improved.

• ETRI Journal
• /
• 제36권4호
• /
• pp.635-642
• /
• 2014

Three-dimensional integrated circuits (3D ICs) implement heterogeneous systems in the same platform by stacking several planar chips vertically with through-silicon via (TSV) technology. 3D ICs have some advantages, including shorter interconnect lengths, higher integration density, and improved performance. Thermal-aware design would enhance the reliability and performance of the interconnects and devices. In this paper, we propose thermal-aware floorplanning with min-cut die partitioning for 3D ICs. The proposed min-cut die partition methodology minimizes the number of connections between partitions based on the min-cut theorem and minimizes the number of TSVs by considering a complementary set from the set of connections between two partitions when assigning the partitions to dies. Also, thermal-aware floorplanning methodology ensures a more even power distribution in the dies and reduces the peak temperature of the chip. The simulation results show that the proposed methodologies reduced the number of TSVs and the peak temperature effectively while also reducing the run-time.

• 대한원격탐사학회지
• /
• 제16권2호
• /
• pp.145-156
• /
• 2000

Clustering is the process of partitioning a data set into meaningful clusters. As the data to process increase, a laster algorithm is required than ever. In this paper, we propose a clustering algorithm to partition a multispectral remotely sensed image data set into several clusters using a hash search algorithm. The processing time of our algorithm is compared with that of clusters algorithm using other speed-up concepts. The experiment results are compared with respect to the number of bands, the number of clusters and the size of data. It is also showed that the processing time of our algorithm is shorter than that of cluster algorithms using other speed-up concepts when the size of data is relatively large.