• 제목/요약/키워드: Convolution method

검색결과 588건 처리시간 0.026초

디지털 영상의 퍼지시스템 표현을 이용한 Edge 검출방법 (An edge detection method for gray scale images based on their fuzzy system representation)

  • 문병수;이현철;김장열
    • 한국지능시스템학회논문지
    • /
    • 제11권6호
    • /
    • pp.454-458
    • /
    • 2001
  • 이 논문에서는 디지털 영상의 퍼지 시스템 표현으로부터 유도된 Edge 검출 알고리듬에 대하여 기술한다. 이 알고리듬은 Gradient을 기반으로 한 것으로 Convolution Kernel이 기존의 Roberts, Prewitt 또는 Sobel등이 제안한 Gradient Kernel과 다른 새로운 것이다. 사용한 퍼지시스템은 디지털 영상을 근사적으로 표현한 Bicubic Spline 함수를 퍼지시스템 화한것으로서 2차 도함수가 연속이기 때문에 Gradient나 Laplacian 연산이 가능하다. Grid 점들에서 이 함수의 Gradient는 두 개의 축 방향으로 각각 한개의 3$\times$3행렬과 영상과의 Covolution에 의하여 산출됨을 보였으며 이를 이용하여 검출된 Edge들은 기존의 다른 방법을 사용하여 검출된 Edge 영상보다 훨씬 선명함을 확인하였다. 이 알고리듬 적용사례 2개에 대한 기술에 포함되어 있다.

  • PDF

변형 가능한 컨볼루션 네트워크와 지식증류 기반 위성 영상 빌딩 분할 (Satellite Building Segmentation using Deformable Convolution and Knowledge Distillation)

  • 최근훈;이응빈;최병인;이태영;안종식;손광훈
    • 한국멀티미디어학회논문지
    • /
    • 제25권7호
    • /
    • pp.895-902
    • /
    • 2022
  • Building segmentation using satellite imagery such as EO (Electro-Optical) and SAR (Synthetic-Aperture Radar) images are widely used due to their various uses. EO images have the advantage of having color information, and they are noise-free. In contrast, SAR images can identify the physical characteristics and geometrical information that the EO image cannot capture. This paper proposes a learning framework for efficient building segmentation that consists of a teacher-student-based privileged knowledge distillation and deformable convolution block. The teacher network utilizes EO and SAR images simultaneously to produce richer features and provide them to the student network, while the student network only uses EO images. To do this, we present objective functions that consist of Kullback-Leibler divergence loss and knowledge distillation loss. Furthermore, we introduce deformable convolution to avoid pixel-level noise and efficiently capture hard samples such as small and thin buildings at the global level. Experimental result shows that our method outperforms other methods and efficiently captures complex samples such as a small or narrow building. Moreover, Since our method can be applied to various methods.

시간적분형 운동방정식을 바탕으로 한 동적 응력확대계수의 계산 (Numerical Computation of Dynamic Stress Intensity Factors Based on the Equations of Motion in Convolution Integral)

  • 심우진;이성희
    • 대한기계학회논문집A
    • /
    • 제26권5호
    • /
    • pp.904-913
    • /
    • 2002
  • In this paper, the dynamic stress intensity factors of fracture mechanics are numerically computed in time domain using the FEM. For which the finite element formulations are derived applying the Galerkin method to the equations of motion in convolution integral as has been presented in the previous paper. To assure the strain fields of r$^{-1}$ 2/ singularity near the crack tip, the triangular quarter-point singular elements are imbedded in the finite element mesh discretized by the isoparametric quadratic quadrilateral elements. Two-dimensional problems of the elastodynamic fracture mechanics under the impact load are solved and compared with the existing numerical and analytical solutions, being shown that numerical results of good accuracy are obtained by the presented method.

사천만에서 취송류의 신속예측에 관한 연구 (Study on the Fast Predication of the Wind-Driven Current in the Sachon Bay)

  • 최석원;조규대;김동선
    • 한국환경과학회지
    • /
    • 제8권3호
    • /
    • pp.309-318
    • /
    • 1999
  • In order to fast predict the wind-driven current in a small bay, a convolution method in which the wind-driven current can be generated only wih the local wind is developed and applied in the Sachon Bay. The root mean square(rms) ratio defined as the ratio of the rms error to the rms speed is 0.37. The rms ratio is generally less than 0.2, except for all the mouths of Junju Bay and Namhae-do and in the region between Saryang Island and Sachon. The spatial average of the recover rate of kinetic energy(rrke) is 87%. Thus, the predicted wind-driven current by the convolution model is in a good agreement with the computed one by the numerical model. The raio of the difference between observed residual current (Vr) and predicted wind-driven current (Vc) to a residual current, that is, (Vr-Vc)/Vr shows 56%, 62% at 2 moorings in the Sachon Bay.

  • PDF

Discrete singular convolution method for bending analysis of Reissner/Mindlin plates using geometric transformation

  • Civalek, Omer;Emsen, Engin
    • Steel and Composite Structures
    • /
    • 제9권1호
    • /
    • pp.59-75
    • /
    • 2009
  • In this study, a simple approach for bending analysis of Reissner-Mindlin plates is presented using the four-node quadrilateral domain transformation based on discrete singular convolution. In the proposed approach, irregular physical domain is transformed into a rectangular domain by using the geometric coordinate transformation. The DSC procedures are then applied to discrete the governing equations and boundary conditions. The accuracy of the proposed method is verified by comparison with known solutions obtained by other numerical or analytical methods. Results for Reissner-Mindlin plates show a satisfactory agreement with the analytical and numerical solutions.

자동차 배기계에 대한 음질 향상 (Improving Sound Quality of the Exhaust System Using Convolution Analysis)

  • Yunseon Ryu;Kim, Yoon-Seok
    • 한국소음진동공학회:학술대회논문집
    • /
    • 한국소음진동공학회 2002년도 춘계학술대회논문집
    • /
    • pp.1148-1150
    • /
    • 2002
  • The exhaust system could be a dominant acoustical source in the passengers vehicle. It would be very important to obtain the acoustically good exhaust system, in order to control the cabin interior sound in automotive. In order to obtain the acoustically good exhaust system in automotive, many kinds of exhaust system should be measured, and simultaneously those results should be compared by the sound quality parameters. In this paper, in order to develop the methodology determining sound quality parameters, acoustic simulator is introduced, combining the time domain analysis and convolution analysis. As an example to verify the reliability of this method, several kinds of measurements are carried out, and the acoustically good exhaust system is selected, based on this proposed method.

  • PDF

Stability and non-stationary vibration analysis of beams subjected to periodic axial forces using discrete singular convolution

  • Song, Zhiwei;Li, Wei;Liu, Guirong
    • Structural Engineering and Mechanics
    • /
    • 제44권4호
    • /
    • pp.487-499
    • /
    • 2012
  • Dynamic instability of beams subjected to periodic axial forces is studied using the discrete singular convolution (DSC) method with the regularized Shannon's delta kernel. The principal regions of dynamic instability under different boundary conditions are examined in detail, and the non-stationary vibrations near the stability-instability critical regions have been investigated. It is found that the results obtained by using the DSC method are consistent with the analytical solutions, which shows that the DSC algorithm is suitable for the problems considered in this study. It was found that there is a narrow region of beat vibration existed in the vicinity of one side (${\theta}/{\Omega}$ > 1) of the boundaries of the instable region for each condition.

Performance Analysis of OFDM System Considering Carrier Frequency Offset in Wireless LAN Channel Environment

  • Kim, Ji-Woong;Kang, Heau-Jo;Lee, Kwon-Hyun
    • Journal of information and communication convergence engineering
    • /
    • 제2권1호
    • /
    • pp.1-4
    • /
    • 2004
  • In this paper, We analyzes how a synchronization error affects receiving system when using OFDM(Orthogonal Frequency Division Multiplexing) transmission method in wireless LAN channel environment in which we can efficiently transmit wide-band information data. As a performance improvement method, performance distortion can be improved by applying convolution coding. As a result, in OFDM system, we could see that the higher a frequency offset is, the worse performance will be, and we could see that there was performance improvement by applying convolution coding in OFDM system in order to reach (BER=$10^{-3}$). However, when we use 64QAM (64Quadrature Amplitude Modulation), there was a huge influence between carriers by frequency offset at 0.05, 0.1.

다양한 포맷변환을 지원하는 Transcoder의 개선된 Cubic Convolution Scaler (Modified Cubic Convolution Scaler for Multiformat Conversion in a Transcoder)

  • 유영조;서주헌;한종기
    • 한국통신학회논문지
    • /
    • 제32권9C호
    • /
    • pp.867-880
    • /
    • 2007
  • 본 논문에서는 임의의 배율로 디지털 영상을 확대와 축소하는 개선된 cubic convolution 보간법을 제안한다. 제안하는 공간해상도조정 방법은 압축된 HD 비트스트림이나 SD 비트 스트림을 transcoding시키는 과정에서 다양한 형식의 해상도를 조정하기위해 사용된다. Transcoder등 다양한 응용분야에서 영상의 화질은 유지하면서 크기를 조정하는 것은 매우 중요한 기술이다. 해상도조정은 원본 디지털 데이터를 연속함수로 변환하는 단계와 새로운 샘플링간격에 맞게 재표본화하는 단계, 이렇게 두 단계로 구분된다. 우리는 원본영상과 해상도조정 된 영상사이의 관계를 고려하여 보간 필터의 kernel 개선에 초점을 맞췄다. 본 논문에서는 MPEG 표준에서 고려하는 다양한 영상포맷들간 해상도를 변환시키기 위한 기술을 제안한다. 이 방법은 기존의 방법과 비교하였을 때, 정보 손실을 최소화하여 고화질의 해상도 변환 기능을 가능하게 한다.

그레이스케일 영상의 병렬가산 컨볼루션 알고리즘 (Parallel-Addition Convolution Algorithm in Grayscale Image)

  • 최종호
    • 한국정보전자통신기술학회논문지
    • /
    • 제10권4호
    • /
    • pp.288-294
    • /
    • 2017
  • 최근들어 CNN(Convolutional Neural Network)을 이용한 딥러닝 기술이 영상인식 등의 분야에서 널리 활용되고 있다. CNN에서 승산과 가산으로 수행되는 컨볼루션 처리는 단순한 연산이지만 하드웨어로 구현하는 데 문제가 되는 것은 승산을 수행하는데 필요한 계산시간이다. 컴퓨팅 파워의 사용에 문제가 없는 응용분야에서는 문제가 되지 않지만 임베디드용 딥러닝 시스템 등의 구현을 위한 하드웨어 칩설계에서는 많은 제한이 있다. 따라서 본 논문에서는 그레이스케일 영상을 2진영상의 중첩으로 표현한 후, 병렬로 가산만을 이용하여 컨볼루션을 수행하는 병렬가산 알고리즘을 제안하였다. 본 논문에서 새롭게 제안한 알고리즘의 유용성을 확인하기 위한 실험을 통해 처리시간의 감소가 가능한 병렬가산 방식으로 컨볼루션을 수행할 수 있음을 확인하였다.