• Journal of KIISE:Software and Applications
• /
• v.36 no.11
• /
• pp.960-966
• /
• 2009

A manifold is used to represent a relationship between high-dimensional data samples in low-dimensional space. In human pose estimation, it is created in low-dimensional space for processing image and 3D body configuration data. Manifold learning is to build a manifold. But it is vulnerable to silhouette variations. Such silhouette variations are occurred due to view-change, person-change, distance-change, and noises. Representing silhouette variations in a single manifold is impossible. In this paper, we focus a silhouette variation problem occurred by view-change. In previous view invariant pose estimation methods based on manifold learning, there were two ways. One is modeling manifolds for all view points. The other is to extract view factors from mapping functions. But these methods do not support one by one mapping for silhouettes and corresponding body configurations because of unsupervised learning. Modeling manifold and extracting view factors are very complex. So we propose a method based on triple manifolds. These are view manifold, pose manifold, and body configuration manifold. In order to build manifolds, we employ biased manifold learning. After building manifolds, we learn mapping functions among spaces (2D image space, pose manifold space, view manifold space, body configuration manifold space, 3D body configuration space). In our experiments, we could estimate various body poses from 24 view points.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.669-680
• /
• 2001

Genetic Algorithms are optimization algorithm that mimics biological evolution to solve optimization problems. Genetic algorithms provide an alternative to traditional optimization techniques by using directed random searches to locate optimal solutions in complex fitness landscapes. Hybrid genetic algorithm that is combined with local search called learning can sustain the balance between exploration and exploitation. The genetic traits that each individual in the population learns through evolution are transferred back to the next generation, and when this learning is combined with genetic algorithm we can expect the improvement of the search speed. This paper proposes a genetic algorithm based Cellular Learning with accelerated learning capability for function optimization. Proposed Cellular Learning strategy is based on periodic and convergent behaviors in cellular automata, and on the theory of transmitting to offspring the knowledge and experience that organisms acquire in their lifetime. We compared the search efficiency of Cellular Learning strategy with those of Lamarckian and Baldwin Effect in hybrid genetic algorithm. We showed that the local improvement by cellular learning could enhance the global performance higher by evaluating their performance through the experiment of various test bed functions and also showed that proposed learning strategy could find out the better global optima than conventional method.

• Korean Journal of Remote Sensing
• /
• v.36 no.2_1
• /
• pp.139-153
• /
• 2020

Using a computationalfluid dynamics(CFD) model, thisstudy evaluated the representativeness of an air quality monitoring system (AQMS) in an urban area and presented a methodology to determine the suitable AQMS locations for specific purposes. For this, we selected a 1.6 km × 1.6 km area around the Eunpyeong-gu AQMS (AQMS 111181) as a target area. We conducted simulationsfor two emission scenarios (scenario one: air pollutants transported from inflow boundaries, scenario two: air pollutants emitted from roads). Urban airflows were markedly influenced by mountainous terrain located in the northeast and southeast of the target area, and complicated airflow patterns occurred around the buildings. The distributions of air pollutants were dependent on the terrain (mountain) in scenario one, but the road location and building height in scenario 2. We evaluated whether the AQMS could represent the air quality in the target area based on the simulations for two scenarios. The concentrations simulated at the AQMS were similar in magnitude to the layer mean concentrations, which indicated good representativeness for the air quality in the target area. We also suggested which locations were suitable for different measurement purposes (hot spots, clean zones, average zones, shelter zones, equi-background zones).

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.22 no.6
• /
• pp.600-606
• /
• 2016

Seafarers are often placed in circumstances that require emergency evacuations due to various causes, including ship collisions, sinking, stranding, and fires. Achieving shorter evacuation time is an important factor in increasing the survival rate during these circumstances, but the narrow and complicated structure of ships is an obstacle when it comes to executing a quick evacuation. Also, unpredictable restrictions may be imposed by bad sea or weather. In this study, various experiments were conducted with sailors currently on board ships in order to examine factors that increase evacuation time. The data was then and analyzed. Evacuation time was measured by dividing crews into groups: sailors that were given an explanation of the ship's structure and those that were not. Furthermore, the visibility range was divided into 0 m, 3 m, and 5 m. The results indicated that, having an explanation of the ship structure did not have much of an effect on evacuation time but visibility conditions led to an increase in evacuation time with a maximum of 2.5 to 2.6 times longer when the visible distance was 5 m, 0 m and 3 m. Therefore, ensuring a visible distance of over 5 m was determined to be the most important factor for reducing evacuation time. In the future, effort should be made to ensure a greater visible distance to improve the survival rate of seafarers and passengers on board ships that encounter incidents.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.9_10
• /
• pp.429-440
• /
• 2008

As embedded system becomes more complex, software development becomes more important in the entire design process. Most embedded applications consist of multi -tasks, that are executed in parallel. So, dataflow model that expresses concurrency naturally is preferred than sequential programming language to develop multitask software. For the execution of multitasking codes, operating system is essential to schedule multi-tasks and to deal with the communication between tasks. But, it is needed to execute multitasking code without as when the target hardware platform cannot execute as or target platforms are candidates of design space exploration, because it is very costly to port as for all candidate platforms of DSE. For this reason, we propose the serialized multitasking code generation technique from dataflow specification. In the proposed technique, a task is specified with dataflow model, and generated as a C code. Code generation consists of two steps: First, a block in a task is generated as a separate function. Second, generated functions are scheduled by a multitasking scheduler that is also generated automatically. To make it easy to write customized scheduler manually, the data structure and information of each task are defined. With the preliminary experiment of DivX player, it is confirmed that the generated code from the proposed framework is efficiently and correctly executed on the target system.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.699-706
• /
• 2014

As XML documents are getting bigger and more complex, a keyword-based search method that does not require structural information is needed to search these large XML documents. In order to use this method, not only all keywords expressed as nodes in the XML document must be labeled for indexing but also structural information should be well represented. However, the existing labeling methods either have very simple information of XML documents for index or represent the structural information which is difficult to deal with the increase of XML documents' size. As the size of XML documents is getting larger, it causes either the poor performance of keyword search or the exponential increase of space usage. In this paper, we present the Repetitive Prime Labeling Scheme (RPLS) in order to improve the problem of the existing labeling methods for keyword-based search of large XML documents. This method is based on the existing prime number labeling method and allows a parent's prime number to be used at a lower level repeatedly so that the number of prime numbers being generated can be reduced. Then, we show an experimental result of the comparison between our methods and the existing methods.

• Journal of Korean Society of Environmental Engineers
• /
• v.34 no.2
• /
• pp.86-90
• /
• 2012

The gaseous product of nitrogen cycle, nitrous oxide ($N_2O$) is a potent greenhouse gas whose Global Warming Potential (GWP) is about 300 times greater than $CO_2$. The dynamics of $N_2O$ emission are controlled by such environments and soil conditions. The main aim of this study is to investigate variations of $N_2O$ emission and its controlling factors with different land-use patterns in Haean basin. A forest, a radish field and a rice paddy were selected as three different land-use patterns. Their $N_2O$ emissions were measured every month during a growing season. We also collected soil samples with seasons and analyzed soil characteristics including inorganic nitrogen content. $N_2O$ emission was greatest at the radish field likely due to anthropogenic nitrogen addition by fertilization. Soils of forest and rice paddy also contained inorganic nitrogen originated from organic matter. However, the spatial variation was great and it looks that nitrogen cycle and $N_2O$ production were slower than that of radish field. Intensive observation and control of fertilization would be requiredto adjust $N_2O$ emission from agriculture soils.

• The KIPS Transactions:PartA
• /
• v.16A no.6
• /
• pp.411-418
• /
• 2009

Vector quantization algorithm based on fuzzy clustering has been widely used in the field of data compression since the use of fuzzy clustering analysis in the early stages of a vector quantization process can make this process less sensitive to its initialization. However, the process of fuzzy clustering is computationally very intensive because of its complex framework for the quantitative formulation of the uncertainty involved in the training vector space. To overcome the computational burden of the process, this paper introduces an array architecture for the implementation of fuzzy vector quantization (FVQ). The arrayarchitecture, which consists of 4,096 processing elements (PEs), provides a computationally efficient solution by employing an effective vector assignment strategy during the clustering process. Experimental results indicatethat the proposed parallel implementation providessignificantly greater performance and efficiency than appropriately scaled alternative array systems. In addition, the proposed parallel implementation provides 1000x greater performance and 100x higher energy efficiency than other implementations using today's ARMand TI DSP processors in the same 130nm technology. These results demonstrate that the proposed parallel implementation shows the potential for improved performance and energy efficiency.

• Tunnel and Underground Space
• /
• v.16 no.6 s.65
• /
• pp.506-525
• /
• 2006

For the design of large scale rock slope which has complex formations and geological structures, generally, insufficiency of geotechnical investigations and laboratory tests are the main factors of slope failures doling construction. In such case, remedial measures to stabilize slope should be selected and applied through reliable investigations and analysis considering the geotechnical characteristics. The rock slope of this study, one of the largest cut slopes in Korea with a length of 520.0 m and maximum height of 122.0 m consists of metasedimentary rocks. And a case study on the causes of large-scale rock slope failure was carried out by analysis of landslides history and site investigations during construction. When the slope with the original design slope of 0.7: 1.0 (H:V) was partially constructed, the slope failure was occurred due to the factors such as poor conditions of rocks (weathered zone, coaly shale and fault shear zone), various discontinuities (joints, foliations and faults), severe rain storm and so on. The types of failures were rockfall, circular failure, wedge failure and the combination of these types. So, the design of slope was changed three times to ensure long-term slope stability. This paper is intended to be a useful reference for analyzing and estimating the stability of rock slopes whose site conditions are similar to those of this study site such as geological structures and geotechnical properties.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.1727-1731
• /
• 2008

수자원의 효율적인 관리를 위해서는 신뢰성 있는 유량자료의 획득이 대단히 중요하다. 우리나라는 양질의 유량자료를 획득하기 위해 매년 많은 시간과 돈을 투자하고 있으나 자료의 질적인 면에서 만족할 만한 성과를 얻지 못하고 있다. 현재까지 우리나라의 유량자료는 댐의 수문자료와 수량관리 부처인 건교부에서 운영하는 수위표 지점의 수위-유량곡선에서 산출된 자료에 의존하고 있다. 그러나 수위-유량 관계식을 보정하기 위한 유량측정사업이 지속적이지 못하며, 이 관계식은 유량이 적은 저수기 및 갈수기에는 부정확하다는 한계가 있다. 또한, 국립환경과학원 낙동강물환경연구소에서 오염총량관리를 위한 낙동강수계 유량측정사업을 실시하고 있지만, 목적은 낙동강수계의 오염총량관리 단위유역 말단 47개 지점에서 유량측정을 효율적으로 실시하여 수질정책의 기초자료를 제공하는데 있다. 이 자료 역시 오염총량관리를 위하여 유량측정을 실시하여 수자원의 효율적인 관리를 위한 일 유량을 알 수가 없는 한계점을 가지고 있다. 따라서 저수기 및 갈수기에 수질정책의 기초자료를 제공하기 위해서 하천을 포함한 유역의 정확한 강우-유출특성의 파악이 필요하다. 그러나 강우-유출특성 또한 유역 내 강우의 시 공간적 분포가 다르며 그 자가 비선형성이 강하고 여러 변동성을 포함하므로, 강우로부터 하천의 유출량의 정확한 해석이 불가능하다. 그러나 최근 인공지능 분야에서 신호처리, 지능제어 및 패턴인식 등의 수단으로 사용되고 있는 신경망은 학습이라는 최적화 과정을 통해 입력과 출력으로 구성되는 하나의 시스템을 비선형적으로 구축할 수 있으며 이러한 이점을 활용하여 수자원 분야에서 다양하게 적용되고 있다. 본 연구의 목적은 강우-유출자료 및 댐 방류량 자료의 비선형적인 특정을 가장 잘 반영할 수 있는 신경망모형을 적용하여 수질정책의 기초자료를 제공하기 위하여 신뢰성 있는 유량자료를 산정하는 모형을 개발하는 것이다. 이를 위해서 낙동강물환경연구소에서 오염총량관리를 위한 낙동강수계 유량측정 지점 상류의 댐 방류량의 일 방류량자료와 강우자료를 입력 자료로 하여 유량을 예측할 수 있는 유량예측 신경망 모형 FFBN(Flow Forecasting By Neural)을 개발하였다. 그리고 입력 자료로서 장기유출모형인 SWAT의 모의결과를 입력 자료로 추가한 FFBNS(Flow Forecasting By Neural and SWAT)을 개발하였다. 신경망 모형의 구조는 입력층과 출력층 사이에 하나의 은닉층이 존재하는 다층 신경망으로 구성하였으며, 학습단계에서는 오류 역전파 알고리듬 학습방법 중 모멘텀법을 사용하였다. 예측된 유출량을 실측치와의 비교를 위하여 낙본D지점과 낙본 E지점에 대하여 $2005{\sim}2006$년까지의 모의 결과를 낙동 수위측정지점과 구미 수위측정지점의 실측치 통하여 복잡한 비선형성을 가지는 유출 시계열 자료에 대한 효과적인 최적의 신경망모델을 개발하여 유량을 예측하고 적용 가능성을 검토하고자 한다. 모의 결과는 수질정책의 기초자료 제공에 기여할 수 있을 것으로 판단된다.