• 제목/요약/키워드: variable feature

검색결과 388건 처리시간 0.023초

가변 축척 매개변수를 가진 변형 확률적 경사도 기반 필터의 해석 (Analysis of a Modified Stochastic Gradient-Based Filter with Variable Scaling Parameter)

  • 김해정
    • 한국통신학회논문지
    • /
    • 제31권12C호
    • /
    • pp.1280-1287
    • /
    • 2006
  • 본 논문은 변형 확률적 경사도 기반 (MSGB) 필터를 제안하고 그 필터가 최적화 문제에 대한 해가 될 수 있음을 보여준다. 갱신항으로 첨가된 가변적 축척 매개변수를 가진 비선형 적응 필터인 MSGB 필터의 특성을 분석 한다. 가변 매개변수의 MSGB 필터는 가변 축척 매개변수를 통하여 고정 매개변수의 MSGB 필터와 무매개변수의 MSGB 필터를 연결하는 역할을 한다. 그 안정성 영역과 오조정량도 살펴본다. 시스템 식별에 응용하여 컴퓨터 모의실험을 실행하여 MSGB 필터의 개선된 성능 특성을 보여준다.

DWT 영역에서의 주파수 정보를 활용한 가변 윈도우 기반의 스테레오 정합 알고리즘 (A New Stereo Matching Algorithm based on Variable Windows using Frequency Information in DWT Domain)

  • 서영호;구자명;김동욱
    • 한국정보통신학회논문지
    • /
    • 제16권7호
    • /
    • pp.1437-1446
    • /
    • 2012
  • 본 논문에서는 스테레오 카메라 환경에서 고속으로 깊이 정보를 얻기 위한 응용분야에 적합한 스테레오 정합 기법을 제안하고자 한다. 이러한 조건을 만족하기 위해서 DWT 영역에서의 주파수 정보와 가변 정합창을 이용하는 적응적인 스테레오 정합 기법을 제안한다. 공간 영역에서 영상의 국부적인 특성을 분석하여 정합창의 크기를 결정하고, 주파수 영역에서 영상의 주파수 특성을 분석하여 정합창의 형태 및 스케일링 요소를 결정한다. 주파수 영역에 대한 정보를 이용하기 위해서 로컬 DWT와 전역 DWT를 활용하는 기법을 모두 적용하였다. 본 논문은 스테레오 정합을 위한 제안한 기법은 유사한 수준의 복잡도에서 고정 정합창 기반의 기법과 비교할 때 PSNR이 향상되는 것을 확인하였다.

Mechanics of a variable damping self-centering brace: Seismic performance and failure modes

  • Xie, Xing-Si;Xu, Long-He;Li, Zhong-Xian
    • Steel and Composite Structures
    • /
    • 제31권2호
    • /
    • pp.149-158
    • /
    • 2019
  • The force-deformation behavior, strain distribution and failure modes of a variable damping self-centering brace (VD-SCB) are theoretically analyzed, experimentally studied, and numerically simulated to guide its design. The working principle of the brace is explained by describing the working stages and the key feature points of the hysteretic curve. A large-scale brace specimen was tested under different sinusoidal excitations to analyze the recentering capability and energy dissipation. Results demonstrate that the VD-SCB exhibits a full quasi-flag-shaped hysteretic response, high ultimate bearing capacity, low activation force and residual deformation, and excellent recentering and energy dissipation capabilities. Calculation equations of the strain distribution in different parts of the brace are proposed and are compared with the experimental data and simulated results. The developments of two failure modes are compared. Under normal circumstances, the brace fails due to the yielding of the spring blocking plates, which are easily replaced to restore the normal operating conditions of the brace. A brief description of the design procedure of the brace is proposed for application.

초음파 검사 기반의 용접결함 분류성능 개선에 관한 연구 (Performance Comparison of Neural Network Algorithm for Shape Recognition of Welding Flaws)

  • 김재열;윤성운;김창현;송경석;양동조
    • 한국공작기계학회:학술대회논문집
    • /
    • 한국공작기계학회 2004년도 춘계학술대회 논문집
    • /
    • pp.287-292
    • /
    • 2004
  • In this study, we made a comparative study of backpropagation neural network and probabilistic neural network and bayesian classifier and perceptron as shape recognition algorithm of welding flaws. For this purpose, variables are applied the same to four algorithms. Here, feature variable is composed of time domain signal itself and frequency domain signal itself, Through this process, we confirmed advantages/disadvantages of four algorithms and identified application methods of few algorithms.

  • PDF

10년 동안 성장한 거대 폐국균종 치험 1례 (Growing Pulmonary Aspergilloma for 10 Years - One Case Report -)

  • 심성보
    • Journal of Chest Surgery
    • /
    • 제24권8호
    • /
    • pp.814-818
    • /
    • 1991
  • Infection with the ubiquitous fungus Aspergillus is uncommon. This organism has a low pathogenicity for man. In most patients, pulmonary Aspergilloma is the most common form of disease, and Aspergillus fumigatus is the common species. The natural history of the pulmonary aspergilloma is variable. This report reviews the feature of growing pulmonary aspergilloma for 10 years[from diameter 3cm to diameter 9cm]. This 44-year-old male patient suffered from recurrent hemoptysis for 10 years. This patient was taken removal of aspergilloma and cavernoplasty of left upper lobe. Post-operation course was uneventful.

  • PDF

용접결함의 형상인식을 위한 신경회로망 알고리즘의 성능 비교 (Performance Comparison of Neural Network Algorithm for Shape Recognition of Welding Flaws)

  • 김재열;심재기;이동기;김창현;송경석;양동조
    • 한국공작기계학회:학술대회논문집
    • /
    • 한국공작기계학회 2003년도 추계학술대회
    • /
    • pp.271-276
    • /
    • 2003
  • In this study, we compared backpropagation neural network(BPNN) with probabilistic neural network(PNN) as shape recognition algorithm of welding flaws. For this purpose, variables are applied the same to two algorithm. Here, feature variable is composed of time domain signal itself and frequency domain signal itself, Through this process, we comfirmed advantages/disadvantages of two algorithms and identified application methods of two algorithms.

  • PDF

초음파 비파괴 검사기법에 의한 용접결함 분류성능 비교 (Performance Comparison of Welding Flaws Classification using Ultrasonic Nondestructive Inspection Technique)

  • 김재열;유신;김창현;송경석;양동조;김유홍
    • 한국공작기계학회:학술대회논문집
    • /
    • 한국공작기계학회 2004년도 추계학술대회 논문집
    • /
    • pp.280-285
    • /
    • 2004
  • In this study, we made a comparative study of backpropagation neural network and probabilistic neural network and bayesian classifier and perceptron as shape recognition algorithm of welding flaws. For this purpose, variables are applied the same to four algorithms. Here, feature variable is composed of time domain signal itself and frequency domain signal itself. Through this process, we comfirmed advantages/disadvantages of four algorithms and identified application methods of four algorithms.

  • PDF

비젼을 이용한 SMD 부품의 위치 및 자세 계측 (High accuracy measurement of the position and orientation of SMD VR by Computer Vision)

  • 김병엽;송재용;한창수;박종현;이영민
    • 한국정밀공학회:학술대회논문집
    • /
    • 한국정밀공학회 1995년도 춘계학술대회 논문집
    • /
    • pp.371-376
    • /
    • 1995
  • Computer vision is applicated to measure the position and orientation of the SMD on 8mm Camcoder PCB and advanced image processing algorithms for high accuracy and real time processing are proposed. Illumination conditions are optimized for the best image formation and a set of LEDs is used as economic illuminator, which is regarded as a summation of many point sources. Conctete optical system is constructed and the performance of the proposed algorithm is verified by several experiments.

  • PDF

디지털 소프트 칼라 교정인쇄를 위한 새로운 색역 사상방법에 관한 연구 (A study on the new gamut mapping method for digital soft color proofing)

  • 송경철;강상훈
    • 한국인쇄학회:학술대회논문집
    • /
    • 한국인쇄학회 2002년도 춘계 학술발표회
    • /
    • pp.10-18
    • /
    • 2002
  • On the process of cross-media color reproduction, a key feature is the use of gamut mapping techniques to adjust the different color gamuts between displays and printers. Even though a number of GMAs have been published, but there are no method satisfactory enough for more exect color reproduction. In this paper, the gamut mapping methods of nearest point clipping(NPC), centroid clipping (SLIN), straight clipping and cusp clipping(CUSP) were tested and analyzed with color difference, and a new gamut mapping algorithm based on variable anchor point method was proposed.

  • PDF

위치제어를 위한 신경망 2 자유도 PID 제어기 (Two-Degree-of-Freedom PID controller with Neural network for position control)

  • 이정민;하홍곤
    • 융합신호처리학회 학술대회논문집
    • /
    • 한국신호처리시스템학회 2000년도 추계종합학술대회논문집
    • /
    • pp.193-196
    • /
    • 2000
  • ln this paper, we consider to apply of 2-DOF (Degree of Freedom) PID controller at D.C servo motor system. Many control system use I-PD, PIB control system. but the position control system have difficulty in controling variable load and changing parameter We propose neural network 2-DOF PID control system having feature for removal disturbrances and tracking function in the target value point.

  • PDF