• Journal of Computing Science and Engineering
• /
• 제9권4호
• /
• pp.204-213
• /
• 2015

Under the conditional independence assumption among local features, the Naive Bayes Nearest Neighbor (NBNN) classifier has been recently proposed and performs classification without any training or quantization phases. While the original NBNN shows high classification accuracy without adopting an explicit training phase, the conditional independence among local features is against the compositionality of objects indicating that different, but related parts of an object appear together. As a result, the assumption of the conditional independence weakens the accuracy of classification techniques based on NBNN. In this work, we look into this issue, and propose a novel Bayesian network for an NBNN based classification to consider the conditional dependence among features. To achieve our goal, we extract a high-level feature and its corresponding, multiple low-level features for each image patch. We then represent them based on a simple, two-level layered Bayesian network, and design its classification function considering our Bayesian network. To achieve low memory requirement and fast query-time performance, we further optimize our representation and classification function, named relation-based Bayesian network, by considering and representing the relationship between a high-level feature and its low-level features into a compact relation vector, whose dimensionality is the same as the number of low-level features, e.g., four elements in our tests. We have demonstrated the benefits of our method over the original NBNN and its recent improvement, and local NBNN in two different benchmarks. Our method shows improved accuracy, up to 27% against the tested methods. This high accuracy is mainly due to consideration of the conditional dependences between high-level and its corresponding low-level features.

• 조명전기설비학회논문지
• /
• 제22권2호
• /
• pp.63-70
• /
• 2008

본 논문은 현 배전계통계획시스템(DISPLAN)의 지역전력수요예측 알고리즘을 개선하여 다중회귀분석을 이용한 지역전력수요예측 알고리즘을 제시하였다. 지역전력수요예측 알고리즘은 예측의 정확도를 높이기 위해 지역경제와 지역인구와 과거의 판매전력량을 입력변수로 사용하였다. 사례연구로 경북의 경산시, 구미시, 김천시, 영주시를 선정하여 제안한 방법의 정확도를 분석하였다. 사례연구 결과 제안한 방법의 전반적인 정확도는 11.2[%]로 DISPLAN의 12[%]보다 향상되었다. 특히 입력변수의 변동성이 심한 지역의 경우에서 많이 개선되었다. 제안된 방법은 배전계통시스템의 최적투자를 위한 지역전력수요예측에 사용될 것으로 사료된다.

• 한국측량학회지
• /
• 제19권2호
• /
• pp.155-161
• /
• 2001

오늘날 3차원 위치결정에 있어서 위성측량을 활용하여 위치결정이 가능하게 되었다. 최근 국내 지오이드모델의 정밀도가 향상되면서 정밀 수직위치결정에 대한 위성측량의 위치정확도 또한 많이 향상되었다. 그러나 지오이드고를 고려하였을 때, 위성측량의 수직위치결정의 활용은 아직 어려운 실정에 있다. 그러나, 지오이드 변화가 국소지역에서는 미소하게 변화하므로 위성측량 관측에 의한 표고와 정표고는 동일하다고 볼 수 있다. 따라서. 지오이드를 고려치 않은 건설현장에서 GLT(GPS Leveling Technique)의 수직위치결정 측량이 가능하게 되었다. GLT는 재래적 측량장비보다 처리속도가 뛰어난 정표고를 계산하는 방법이었고, 관측방법과 수신기 상태, 현장 상황에 따라 정확도가 다소 변화하게 되는데 본 연구에서는 측점간의 경중률 만을 고려하였다. 지반 침하량 계측에 있어서 레벨을 이용한 재래식 수직위치결정방법을 기준으로 부산대학교내 일정지역과 모델현장에서 지오이드모델을 이용한 위성측량방법과 GLT 방법을 비교 분석하여 위치 정확도 및 시간과 비용에 있어서의 효율성을 검토하였다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제8권1호
• /
• pp.37-44
• /
• 2019

디지털 IT 기술의 발달로 인하여 프린터와 스캐너의 성능이 향상되고 가격이 저렴해지면서 일반인들도 쉽게 접할 수 있게 되었다. 그러나 이에 따른 부작용으로 공문서 및 사문서 위조 등의 범죄들이 쉽게 이루어질 수 있다. 따라서 해당 문서가 어떤 프린터를 사용하여 출력 되었는가를 특정할 수 있다면 수사 범위를 줄이고 용의자를 판별하는데 도움이 된다. 본 논문에서는 프린터 장치 판별을 위하여 딥러닝 모델을 제안한다. 먼저 최근 인식 등에서 범용적으로 활용되는 지역 특징 기반의 컨볼루셔널 뉴널 네트워크를 이용한 프린터 장치 판별 모델을 제안하고, 전역 특징 기반의 처리 과정을 네트워크 모델에 도입함으로 인하여 수렴 속도 및 정확도를 향상한 기법을 제안한다. 제안한 모델의 성능은 8개의 프린터 장치를 활용하여 기존 프린터 판별을 위한 특징 기반 기술과 비교를 수행하였다. 그 결과 제안하는 지역 특징 기반의 모델과 전역 특징 기반의 모델이 각각 97.23% 및 99.98%의 높은 판별 정확도를 달성하였고, 기존 기술들에 비하여 높은 정확도를 갖는 우수성을 보였다.

• Interaction and multiscale mechanics
• /
• 제7권1호
• /
• pp.543-561
• /
• 2014

We present a full energy and force formulation of the quasicontinuum method with non-local and local transition elements. Non-local transition elements are developed to transmit inhomogeneity from the atomistic to the continuum regions. Local transition elements are developed to resolve the mathematical mismatch between non-local atoms and the local continuum. The rationale behind these transition elements is provided by analyzing the energy and force transitions between atoms and continuum under the Cauchy-Born rule. We show that breakdown of the Cauchy-Born rule occurs for slaved atoms of local elements within the cutoff of non-local atoms. The inadequacy of the Cauchy-Born rule at the transition region naturally leads to the need of atomistic treatment of transition slaved and transition representative atoms. Such an atomistic treatment together with a full or cutoff sampling allows non-local transition elements containing these transition entities to transmit inhomogeneity. Different force formulations for transition representative atoms and pure local representative atoms allow the local transition elements to resolve non-local and local mismatches. The method presented herein is validated by force calculations in an unstressed perfect crystal as well as an unrelaxed grain boundary model. A nanoindentation simulation in 3D is conducted to demonstrate the accuracy and efficiency of the proposed method.

• 한국지능시스템학회논문지
• /
• 제23권1호
• /
• pp.18-23
• /
• 2013

In this paper we propose an image enhancement technique based on histogram specification method over local overlapping regions referred as Local Histogram Specification. First, both reference and original images are splitted into local regions that each overlaps half of its adjacent regions and general histogram specification method is used between corresponding local regions of reference and original image. However it produces noticeable boundary effects. Linear weighted image blending method is used to reduce this effect in order to make seamless image and we also proposed new technique dealing with over-enhanced contrast areas. We satisfied with our experimental results that showed better enhancement accuracy and less noise amplifications compared to other well-known image enhancement methods. We conclude that the proposed method is well suited for motion detection systems as a responsible part to overcome sudden illumination changes.

• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.849-858
• /
• 2012

Today, local environmental factors has an influence on our society. Local environmental factors, as well as weather-related natural phenomena, social phenomena are also included. In this paper, numeric factors and categorical factors were analyzed, looking for a local environmental factors affecting the company's sales.Sales model by performing a regression analysis based on this was implemented.Sales model considering the local environment had an accuracy of 88.89%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4534-4555
• /
• 2015

We propose a robust object cataloging method using multiple locally distinct heterogeneous features for aiding image retrieval. Due to challenges such as variations in object size, orientation, illumination etc. object recognition is extraordinarily challenging problem. In these circumstances, we adapt local interest point detection method which locates prototypical local components in object imageries. In each local component, we exploit heterogeneous features such as gradient-weighted orientation histogram, sum of wavelet responses, histograms using different color spaces etc. and combine these features together to describe each component divergently. A global signature is formed by adapting the concept of bag of feature model which counts frequencies of its local components with respect to words in a dictionary. The proposed method demonstrates its excellence in classifying objects in various complex backgrounds. Our proposed local feature shows classification accuracy of 98% while SURF,SIFT, BRISK and FREAK get 81%, 88%, 84% and 87% respectively.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2007년도 정기 학술대회 논문집
• /
• pp.405-410
• /
• 2007

We present a new global/local analysis with the aid of MLS(Moving Least Square)-based finite elements which can handle an arbitrary number of nodes on every element side. It give a great flexibility in constructing finite element meshes at the specified local regions without remeshing. Compared to other type global/local analysis, it does not require any superimposed mesh or need not solve the equilibrium equation twice as well as shows an excellent accuracy. To demonstrate the performance of proposed scheme, we will show several examples in relation to capturing highly local stress field.

• Steel and Composite Structures
• /
• 제20권3호
• /
• pp.693-718
• /
• 2016

This paper presents a finite element procedure based on Bridging multi-scale method (BMM) in order to incorporate the effect of local/cross-sectional deformations (e.g., flange local buckling and web crippling) on the global behaviour of thin-walled members made of fibre-reinforced polymer composite laminates. This method allows the application of local shell elements in critical regions of an existing beam-type model. Therefore, it obviates the need for using computationally expensive shell elements in the whole domain of the structure, which is otherwise necessary to capture the effect of the localized behaviour. Consequently, highly accurate analysis results can be achieved with this method by using significantly smaller finite element model, compared to the existing methods. The proposed method can be used for composite polymer laminates with arbitrary fibre orientation directions in different layers of the material, and under various loading conditions. Comparison with full shell-type finite element analysis results are made in order to illustrate the efficiency and accuracy of the proposed technique.