• 반도체디스플레이기술학회지
• /
• 제15권3호
• /
• pp.67-71
• /
• 2016

This paper addresses a problem of defocus map recovery from single image. We describe a simple effective approach to estimate the spatial value of defocus blur at the edge location of the image. At first, we perform a re-blurring process using Gaussian function with input image, and calculate a gradient magnitude ratio with blurring amount between input image and re-blurred image. Then we get a full defocus map by propagating the blur amount at the edge location. Experimental result reveals that our method outperforms a reliable estimation of depth map, and shows that our algorithm is robust to noise, inaccurate edge location and interferences of neighboring edges within input image.

• Journal of Information Processing Systems
• /
• 제10권2호
• /
• pp.223-239
• /
• 2014

New IIR digital differintegrators (differentiator and integrator) with very minor absolute relative errors are presented in this paper. The digital integrator is designed by interpolating some of the existing integrators. The optimum value of the interpolation ratio is obtained through linear programming optimization. Subsequently, by modifying the transfer function of the proposed integrator appropriately, new digital differentiator is obtained. Simulation results demonstrate that the proposed differintegrator are a more accurate approximation of ideal ones, than the existing differintegrators. Furthermore, the proposed differentiator has been tested in an image processing application. Edges characterize boundaries and are, therefore, a problem of fundamental importance in image processing. For comparison purpose Prewitt, Sobel, Roberts, Canny, Laplacian of Gaussian (LOG), Zerocross operators were used and their results are displayed. The results of edge detection by some of the existing differentiators are also provided. The simulation results have shown the superiority of the proposed approach over existing ones.

• Journal of Information Processing Systems
• /
• 제15권1호
• /
• pp.7-21
• /
• 2019

In order to improve the edge segmentation effect of the level set image segmentation and avoid the influence of the initial contour on the level set method, a saliency level set image segmentation model based on local Renyi entropy is proposed. Firstly, the saliency map of the original image is extracted by using saliency detection algorithm. And the outline of the saliency map can be used to initialize the level set. Secondly, the local energy and edge energy of the image are obtained by using local Renyi entropy and Canny operator respectively. At the same time, new adaptive weight coefficient and boundary indication function are constructed. Finally, the local binary fitting energy model (LBF) as an external energy term is introduced. In this paper, the contrast experiments are implemented in different image database. The robustness of the proposed model for segmentation of images with intensity inhomogeneity and complicated edges is verified.

• ETRI Journal
• /
• 제44권2호
• /
• pp.194-207
• /
• 2022

In the last decade, substantial progress has been achieved in intelligent traffic control technologies to overcome consistent difficulties of traffic congestion and its adverse effect on smart cities. Edge computing is one such advanced progress facilitating real-time data transmission among vehicles and roadside units to mitigate congestion. An edge computing-based deep reinforcement learning system is demonstrated in this study that appropriately designs a multiobjective reward function for optimizing different objectives. The system seeks to overcome the challenge of evaluating actions with a simple numerical reward. The selection of reward functions has a significant impact on agents' ability to acquire the ideal behavior for managing multiple traffic signals in a large-scale road network. To ascertain effective reward functions, the agent is trained withusing the proximal policy optimization method in several deep neural network models, including the state-of-the-art transformer network. The system is verified using both hypothetical scenarios and real-world traffic maps. The comprehensive simulation outcomes demonstrate the potency of the suggested reward functions.

• Journal of information and communication convergence engineering
• /
• 제20권4호
• /
• pp.273-279
• /
• 2022

Owing to the need to establish a cooperative-intelligent transport system (C-ITS) environment in the transportation sector locally and abroad, various research and development efforts such as high-tech road infrastructure, connection technology between road components, and traffic information systems are currently underway. However, the current central control center-oriented information collection and provision service structure and the insufficient road infrastructure limit the realization of the C-ITS, which requires a diversity of traffic information, real-time data, advanced traffic safety management, and transportation convenience services. In this study, a network construction method based on the existing received signal strength indicator (RSSI) selected as a comparison target, and the experimental target and the proposed intelligent edge network compared and analyzed. The result of the analysis showed that the data transmission rate in the intelligent edge network was 97.48%, the data transmission time was 215 ms, and the recovery time of network failure was 49,983 ms.

• KNOM Review
• /
• 제23권1호
• /
• pp.34-42
• /
• 2020

NFV (Network Function Virtualization) 환경에서는 소프트웨어로 구현된 가상 네트워크 기능 (VNF: Virtualized Network Function)을 범용 서버에 설치하는 것으로 네트워크 기능을 제공한다. 네트워크 관리자는 VNF를 네트워크 토폴로지 상 적절한 위치의 서버에 배치하고 상황에 따라 동적으로 관리함으로써, 다양한 네트워크 상황에 대해 신속하고 유연하게 대응할 수 있다. 하지만 여러 네트워크 조건 (서비스 비용 및 품질) 등을 고려하는 것은 매우 복잡하고 어려운 문제이며, 특히 결정된 배치를 실제 NFV 환경에 적용하는 데는 처리 시간이 소요되기 때문에, 최적의 VNF 배치를 위해서는 필요한 자원량을 예측하여 VNF 배치를 결정하는 것이 필요하다. 본 논문에서는 MEC (Multi-access Edge Computing) 토폴로지에서 서비스 요청을 무작위로 생성하여 ILP (Integer Linear Programming) 모델을 통해 시뮬레이션한 결과를 학습데이터로 사용하는 기계학습 모델을 도출한다. 도출된 예측 모델은 5분 이후의 미래 시점에 대해 ILP 솔루션 결과 대비 90% 이상의 정확도를 보였다.

• 정보처리학회논문지B
• /
• 제17B권1호
• /
• pp.63-68
• /
• 2010

본 논문은 스테레오 시각에서 에지 정보를 이용한 유전 알고리즘 기반의 다해상도 스테레오 영상 정합 방법을 제안하고자 한다. 정합 환경을 최적화 문제로 간주하여 유전 알고리즘을 이용하여 해를 찾는다. 비용함수는 스테레오 정합에서 주로 고려할 수 있는 제약 조건으로 구성하였다. 처리의 효율성을 높이기 위해, 영상 피라미드 방벙을 적용하여 최저해상도에서 최초 변위도를 계산한다. 그리고 최초 변위도는 다음 해상도로 전파되고, 보간된 후 변위 정제를 수행한다. 실험을 통해 제안한 방법이 변위 탐색 시간을 감소시킬 뿐만 아니라 정합의 타당성을 보증함을 확인하고자 한다.

• 한국해양공학회지
• /
• 제26권4호
• /
• pp.42-49
• /
• 2012

A numerical method is presented for an out-of-plane structural intensity analysis of rectangular thick plates with arbitrary elastic edge constraints. The method adapts an assumed mode method based on Timoshenko beam functions to obtain the velocities and internal forces needed for a structural intensity analysis. To verify the presented method, the structural intensity of a square thick plate under harmonic force excitation, for which four edges are simply supported, is analyzed, and the result is compared with existing solutions using the assumed mode method based on trigonometric functions. In addition, numerical analyses are carried out for a rectangular-shaped thick plate under harmonic force excitations, of which three edges are simply supported and one edge utilizes an arbitrary elastic edge constraint. These numerical examples show the good accuracy and applicability of the presented method for rectangular thick plates with arbitrary edge constraints.

• 한국강구조학회 논문집
• /
• 제21권4호
• /
• pp.433-442
• /
• 2009

본 연구는 V노치 또는 예리한 균열이 존재하는 N 다변형 단면 입체 실린더에 대한 새로운 3차원 진동 데이터를 제시한다. 본 논문에서는 수학적으로 완전한 대수삼각다항식과 V노치 선단을 따라 존재하는 3차원 응력특이도를 명확히 다루는 허용에지함수와 함께 Ritz방법이 적용된다. 응력특이도를 포함하는 다변형 입체 실린더의 정확한 고유진동수 및 모드형상을 얻기 위해서는 에지함수가 필요하다는 것이 수렴도 분석을 통하여 입증된다.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2014년도 제46회 동계 정기학술대회 초록집
• /
• pp.180.2-180.2
• /
• 2014

Opening a bandgap by forming graphene nanoribbons (GNRs) and tailoring their properties via doping is a promising direction to achieve graphene-based advanced electronic devices. Applying a first-principles computational approach combining density functional theory (DFT) and DFT-based non-equilibrium Green's function (NEGF) calculation, we herein study the structural, electronic, and charge transport properties of boron-nitrogen binary edge doped GNRs and show that it can achieve novel doping effects that are absent for the single B or N doping. For the armchair GNRs, we find that the B-N edge co-doping almost perfectly recovers the conductance of pristine GNRs. For the zigzag GNRs, it is found to support spatially and energetically spin-polarized currents in the absence of magnetic electrodes or external gate fields: The spin-up (spin-down) currents along the B-N undoped edge and in the valence (conduction) band edge region. This may lead to a novel scheme of graphene band engineering and benefit the design of graphene-based spintronic devices.