• 제목/요약/키워드: Partial tuning

검색결과 19건 처리시간 0.025초

Recognition of 3D hand gestures using partially tuned composite hidden Markov models

  • Kim, In Cheol
    • International Journal of Fuzzy Logic and Intelligent Systems
    • /
    • 제4권2호
    • /
    • pp.236-240
    • /
    • 2004
  • Stroke-based composite HMMs with articulation states are proposed to deal with 3D spatio-temporal trajectory gestures. The direct use of 3D data provides more naturalness in generating gestures, thereby avoiding some of the constraints usually imposed to prevent performance degradation when trajectory data are projected into a specific 2D plane. Also, the decomposition of gestures into more primitive strokes is quite attractive, since reversely concatenating stroke-based HMMs makes it possible to construct a new set of gesture HMMs without retraining their parameters. Any deterioration in performance arising from decomposition can be remedied by a partial tuning process for such composite HMMs.

Partial transmission block 제작 시 real block과 MLC를 이용한 방법 중 효율적인 방법에 대한 고찰 (Partial transmission block production for real efficient method of block and MLC)

  • 최지민;박주영;주상규;안종호
    • 대한방사선치료학회지
    • /
    • 제16권2호
    • /
    • pp.19-24
    • /
    • 2004
  • 목적 : Pelvis내에 존재하는 large lesion과 inguinal lymph node를 동시에 치료하고자 할때 femoral head에 과도한 선량을 피할 목적으로 partial transmission block(이하PTB)이 사용되어져 왔다. 그러나 field가 서로 overlap되거나 분리되는 문제를 해결하기 위한 다소 복잡하고 시간도 많이 걸리는 어려움이 있어 본 논문에서는 real block과 MLC를 이용하여 각각 PTB를 제작한 후 몇 가지 비교를 통하여 두 가지 중 실제 임상에 더 효율적으로 사용할 수 있는 방법을 연구하였다. 대상 및 방법 : 실제 치료 환자를 대상으로 디자인 된 PTB를 real block과 MLC를 이용하여 각각 제작한 뒤 아크릴 phantom으로 환자의 두께를 재현하고 치료 시와 동일한 조건으로 노광된 film을 획득하였다. field간에 overlap되는 부분과 분리되는 부분은 block을 미세 조장한 후 다시 촬영하였으며 오차가 1mm이내에 들어 올때까지 junction을 반복 tuning하였다. 두 block을 재현성, 제작 편의성, 제작 시간으로 나누어 비료 분석하였다. 재현성은 5회 반복 측정을 실시하였으며, 제작 편의성 및 제작 시간은 real block과 MLC가 각각 제작 시작 시간부터 완성되는 시점까지에 대하여 측정하였다. 결과 : PTB를 제작함에 있어서 real block과 MLC는 재현성 면에서는 유의할 만한 차이를 보이지 않았다. 그러나 제작 편의성에 있어서는 MLC가 junction tuning을 더 간편하게 수행 할 수 있었으며, 제작 시간 면에 있어서도 MLC가 real block에 비해 약$33\%$정도의 시간 절감 효과가 있음을 알 수 있었다. 결론 : PTB를 제작함에 있어서 real block과 MLC를 이용하는 것이 각각 장단점을 가지고 있으나 real block은 제작 편의성면에서 유연서이 떨어짐으로 각 fie의 junction을 tuning하는데 매우 어려움이 따름과 동시에 비교적 정확한 junction tuning을 시행할 수 있음을 알수 있었다. MLC특성상 발생되는 계단형태의 junction을 보완하여 PTB를 제작한다면 실제 임상에서 훨씬 간편하고 효율성 있는 업무를 수행할 수 있을 것으로 사료된다.

  • PDF

무선 XML 스트림을 위한 색인 기법 (An Index Method for Wireless XML Streams)

  • 정연돈;이지연
    • 한국정보과학회논문지:데이타베이스
    • /
    • 제32권4호
    • /
    • pp.416-428
    • /
    • 2005
  • 본 논문은 무선 정보 시스템 환경에서, 서버가 다수의 클라이언트들에게 무선 방송 기법을 통해 XML 데이타를 스트리밍 서비스할 때 필요한 색인 기법을 제안한다. 제안 하는 색인 방법은 XML 데이타의 스트리밍시 클라이언트들의 접근 시간 및 튜닝 시간을 효과적으로 제어하기 위하여, XML 데이터 및 색인 정보를 부분적으로 반복, 배치하여 스트림을 구성한다. 이를 위하여 트리형태로 표현되는 XML 데이타와 색인 정보를 2-레벨로 구분하여, 색인 및 데이타의 중복 배치 영역을 설정한다. 제안하는 색인 기법의 성능을 접근 시간 및 튜닝 시간 측면에서 분석하여, 분석의 결과로 최적의 레벨 깊이를 결정한다.

A Validation Method for Solution of Nonlinear Differential Equations: Construction of Exact Solutions Neighboring Approximate Solutions

  • Lee, Sang-Chul
    • International Journal of Aeronautical and Space Sciences
    • /
    • 제3권2호
    • /
    • pp.46-58
    • /
    • 2002
  • An inverse method is introduced to construct benchmark problems for the numerical solution of initial value problems. Benchmark problems constructed through this method have a known exact solution, even though analytical solutions are generally not obtainable. The solution is constructed such that it lies near a given approximate numerical solution, and therefore the special case solution can be generated in a versatile and physically meaningful fashion and can serve as a benchmark problem to validate approximate solution methods. A smooth interpolation of the approximate solution is forced to exactly satisfy the differential equation by analytically deriving a small forcing function to absorb all of the errors in the interpolated approximate solution. A multi-variable orthogonal function expansion method and computer symbol manipulation are successfully used for this process. Using this special case exact solution, it is possible to directly investigate the relationship between global errors of a candidate numerical solution process and the associated tuning parameters for a given code and a given problem. Under the assumption that the original differential equation is well-posed with respect to the small perturbations, we thereby obtain valuable information about the optimal choice of the tuning parameters and the achievable accuracy of the numerical solution. Illustrative examples show the utility of this method not only for the ordinary differential equations (ODEs) but for the partial differential equations (PDEs).

ε조절 요소를 가진 부분 상태 궤환 제어기를 이용한 자기부상 시스템의 제어 (Control of Electromagnetic Levitation System using ε-scaling Partial State Feedback Controller)

  • 박규만;최호림
    • 전기학회논문지
    • /
    • 제60권8호
    • /
    • pp.1572-1576
    • /
    • 2011
  • The electromagnetic levitation(EMS) system is one of the well-known nonlinear system because of its nonlinearity and several control techniques have been proposed. We propose an ${\epsilon}$-scaling partial feedback controller for the ball position control of the EMS system. The key feature of our proposed controller is the use of the scaling factor ${\epsilon}$ which provides a function of controller gain tuning along with robustness. In this paper, we show the stability analysis of our proposed controller and the convergence analysis of the state observer in terms of ${\epsilon}$-scaling factor. In addition, the experimental results show the validity of the proposed controller and improved control performance over the conventional PID controller.

Experimental Examination of Multivariable PID Controller Design on Frequency Domain using Liquid Level Process

  • Eguchi, Kazuki;Iwai, Zenta;Mizumoto, Ikuro;Kumon, Makoto
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 2005년도 ICCAS
    • /
    • pp.786-791
    • /
    • 2005
  • This paper is concerned with the examination and evaluation concerning a tuning method of multivariable PID controllers based on partial model matching on frequency domain proposed by authors from practical view point. In this case, PID controller parameters are determined by minimizing the loss function defined by the difference between frequency response of ideal model transfer function and actual frequency response on several frequency points. The purpose of the paper is to examine and evaluate the performance of the method through actual experiments of MIMO liquid level experimental process control equipment.

  • PDF

A Computer-Aided Diagnosis of Brain Tumors Using a Fine-Tuned YOLO-based Model with Transfer Learning

  • Montalbo, Francis Jesmar P.
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제14권12호
    • /
    • pp.4816-4834
    • /
    • 2020
  • This paper proposes transfer learning and fine-tuning techniques for a deep learning model to detect three distinct brain tumors from Magnetic Resonance Imaging (MRI) scans. In this work, the recent YOLOv4 model trained using a collection of 3064 T1-weighted Contrast-Enhanced (CE)-MRI scans that were pre-processed and labeled for the task. This work trained with the partial 29-layer YOLOv4-Tiny and fine-tuned to work optimally and run efficiently in most platforms with reliable performance. With the help of transfer learning, the model had initial leverage to train faster with pre-trained weights from the COCO dataset, generating a robust set of features required for brain tumor detection. The results yielded the highest mean average precision of 93.14%, a 90.34% precision, 88.58% recall, and 89.45% F1-Score outperforming other previous versions of the YOLO detection models and other studies that used bounding box detections for the same task like Faster R-CNN. As concluded, the YOLOv4-Tiny can work efficiently to detect brain tumors automatically at a rapid phase with the help of proper fine-tuning and transfer learning. This work contributes mainly to assist medical experts in the diagnostic process of brain tumors.

유가식 생물반응기에서의 용존산소농도 및 비성장속도의 제어 (Control of dissolved Oxygen Concentration and Specific Growth Rate in Fed-batch Fermentation)

  • 김창겸;이태호;이승철;장용근;장호남
    • 한국미생물·생명공학회지
    • /
    • 제21권4호
    • /
    • pp.354-365
    • /
    • 1993
  • A novel control method with automatic tuning of PID controller parameters has been developed for efficient regulation of dissolved oxygen concentration in fed-batch fermentations of Escherichia coli. Agitation speed and oxygen partial pressure in the inlet gas stream were chosen to be the manipulated variables. A heuristic reasoning allowed improved tuning decisions from the supervision of control performance indices and it coule obviate the needs for process assumptions or disturbance patterns. The control input consisted of feedback and feedforword parts. The feedback part was determined by PID control and the feedforward part is determined from the feed rate. The proportional gain was updated on-line by a set of heuristics rules based on the supervision of three performance indices. These indices were output error covariance, the average value of output error, and input covariance, which were calculated on-line using a moving window. The integral and derivative time constants were determined from the period of output response. The specific growth rate was maintained at a low level to avoid acetic acid accumulation and thus to achieve a high cell density. The specific growthe rate was estimated from the carbon dioxide evolution rate. In fed-batch fermentation, the simutaneous control of dissolved oxygen concentration (at 0.2; fraction of saturated value) and specific growth rate (at 0.25$hr^{-1}$) was satisfactory for the entire culture period in spite of the changes in the feed rate and the switching of control input.

  • PDF

달팽이관 기저막의 이차원적 모델링 (Two-Dimensional Modelling of the Cochlear Basilar Membrane)

  • 장순석
    • 대한의용생체공학회:의공학회지
    • /
    • 제15권4호
    • /
    • pp.439-446
    • /
    • 1994
  • Two-Dimensional modelling of the Cochlear biomechanics is presented in this paper. The Laplace partial differential equation which represents the fluld mechanics of the Cochlea has been transformed into two-dimensional electrical transmission line. The procedure of this transformation is explained in detail. The comparison between one and two dimensional models is also presented. This electrical modelling of the basilar membrane (BM) is clearly useful for the next approach to the further development of active elements which are essenclal in the producing of the sharp tuning of the BM. This paper shows that two-dimension model is qualitatively better than one-dimensional model both in amplitude and phase responses of the BM displacement. The present model is only for frequency response. However because the model is electrical, the two-dimensional transmission line model can be extended to time response without any difficult.

  • PDF

인공장기

  • 민병구
    • 대한의용생체공학회:의공학회지
    • /
    • 제10권2호
    • /
    • pp.112-113
    • /
    • 1989
  • Two-Dimensional modelling of the Cochlear biomechanics is presented in this paper. The Laplace partial differential equation which represents the fluid mechanics of the Cochlea has been transformed into two-dimensional electrical transmission line. The procedure of this transformation is explained in detail. The comparison between one and two dimensional models is also presented. This electrical modelling of the basilar membrane (BM) is clearly useful for the next approach to the further. Development of active elements which are essential in the producing of the sharp tuning of the BM. This paper shows that two-dimension model is qualitatively better than one-dimensional model both in amplitude and phase responses of the BM displacement. The present model is only for frequency response. However because the model is electrical, the two-dimensional transmission line model can be extended to time response without any difficult.

  • PDF