• 한국지능시스템학회논문지
• /
• 제22권2호
• /
• pp.135-140
• /
• 2012

대뇌 상의 mirror neuron system은 시각 정보에 기반한 모방학습 기능을 담당한다. 관측자의 mirror neuron system 영역을 관찰할 때, 행위자가 수행하는 목적성 행위의 전체가 아닌, 부분적으로 가려지거나 보이지 않는 영역을 포함하는 경우에도 해당 영역의 뉴런이 발화되는 과정을 통해 전체 행동의 의도를 유추할 수 있다. 이러한 모방학습 기능을 3D 비전 기반 지능 시스템에 적용하는 것이 본 논문의 목표이다. 본 연구실에서 선행 연구된 스테레오 카메라를 기반으로 획득된 3차원 영상에 대한 복원을 수행한다. 이 때 3차원 입력영상은 부분적으로 가려진 영역을 포함하는 손동작의 순차적 연속영상이다. 복원 결과를 기반으로 가려진 영역을 내포한 행위에 대하여 LK optical flow, unscented Kalman filter를 이용한 특징검출을 수행하고 의도인식의 수행을 위해, Hidden Markov Model을 활용한다. 순차적 입력데이터에 대한 동적 추론 기능은 가려진 영역을 포함한 손동작 인식 수행에 있어 적합한 특성을 가진다. 본 논문에서 제안하는 의도 인식을 위해 선행 연구에서 복원 영상에서의 객체의 윤곽선 및 특징 검출을 시뮬레이션 하였으며, 검출 특징에 대한 시간적 연속 특징벡터를 생성하여 Hidden Markov Model에 적용함으로써, 의도 패턴에 따른 손동작 분류 시뮬레이션을 수행하였다. 사후 확률 값의 형태로 손 동작 분류 결과를 얻을 수 있었으며, 이를 통한 성능의 우수함을 입증하였다.

• Journal of international Conference on Electrical Machines and Systems
• /
• 제3권3호
• /
• pp.325-330
• /
• 2014

Eddy current induction is used in a wide range of electronic devices, for example in detection sensors. Due to the advances in computer hardware and software, the need for 3D computation and system comprehension is a requirement to develop and optimize such devices nowadays. Pure theoretical models are mostly limited to special cases. On the other hand, the classical use of commercial Finite Element (FE) electromagnetic 3D models is not computationally efficient and lacks modeling flexibility or robustness. The proposed approach focuses on: (1) implementing theoretical formulations in 3D (FE) model of a detection device as well as (2) an automatic Volumetric Estimation Method (VEM) developed to selectively model the target finite elements. Due to these two approaches, this model is suitable for parametric studies and optimization of the number, location, shape, and size of PCB receivers in order to get the desired target discrimination information preserving high accuracy with tenfold reduction in computation time compared to commercial FE software.

• 한국콘텐츠학회논문지
• /
• 제14권1호
• /
• pp.24-29
• /
• 2014

본 논문에서는 키넥트 적외선 프로젝터를 통해 깊이를 감지할 수 있는 카메라를 이용하여 사람 움직임을 추적하고 본 논문에서 제안한 몸동작 모델 인식을 통하여 3D 콘텐츠를 제어하는 기법을 제안 한다. 본 논문에서 제안하는 사람의 동작 인식 모델은 사람의 오른팔과 왼팔의 손목, 팔꿈치, 어께 움직임의 거리를 계산하여 좌, 우, 상, 하, 확대, 축소, 선택 등의 7가지 동작 상태를 인식한다. 본 연구에서 제안한 키넥트 기반의 동작 인식 모델은 기존의 접촉식 방식의 인터페이스와 비교할 때 특정센서 또는 장비 부착에 대한 불편함을 없애고 고비용의 하드웨어 시스템을 이용하지 않음으로서 사람의 자연스런 몸동작 이동에 따른 저 비용 3D 콘텐츠 제어 기술을 보여준다.

• Nuclear Engineering and Technology
• /
• 제53권5호
• /
• pp.1446-1453
• /
• 2021

In the past, several experimental investigations aiming at characterizing the natural circulation (NC) behavior in test facilities were carried out. They showed a variety of flow patterns characterized by an inverted U-shape of the NC flow curve versus primary mass inventory. On the other hand, attempts to reproduce such curves using thermal-hydraulic system codes, showed 10-30% differences between the measured and calculated NC mass flow rate. Actually, the used computer codes are generally based upon nodalization using single U-tube representation. Such model may not allow getting accurate simulation of most of the NC phenomena occurring during such tests (like flow redistribution and flow reversal in some SG U-tubes). Simulations based on multi-U-tubes model, showed better agreement with the overall behavior, but remain unable to predict NC phenomena taking place in the steam generator (SG) during the experiment. In the current study, the CATHARE code is considered in order to assess a NC characterization test performed in the four loops PKL facility. For this purpose, four different SG nodalizations including, single and multi-U-tubes, 1D and 3D SG inlet/outlet zones are considered. In general, it is shown that the 1D and 3D models exhibit similar prediction results up to a certain point of the rising part of the inverted U-shape of the NC flow curve. After that, the results bifurcate with, on the one hand, a tendency of the 1D models to over-predict the measured NC mass flow rate and on the other hand, a tendency of the 3D models to under-predict the NC flow rate.

• 한국정밀공학회지
• /
• 제19권11호
• /
• pp.54-64
• /
• 2002

This paper deals with the problem of synthesis of stable and optimal grasps with unknown objects by 3-finger hand. Previous robot grasp research has mainly analyzed with either unknown objects 2-dimensionally by vision sensor or known objects, such as cylindrical objects, 3-dimensionally. As extending the previous work, in this study we propose an algorithm to analyze grasp of unknown objects 3-dimensionally by using vision sensor. This is archived by two steps. The first step is to make a 3-dimensional geometrical model for unknown objects by using stereo matching. The second step is to find the optimal grasping points. In this step, we choose the 3-finger hand which has the characteristic of multi-finger hand and is easy to model. To find the optimal grasping points, genetic algorithm is employed and objective function minimizes the admissible force of finger tip applied to the objects. The algorithm is verified by computer simulation by which optimal grasping points of known objects with different angle are checked.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.1177-1179
• /
• 2003

Recently, the ground-based laser profiler is used for acquisition of 3D spatial information of a rchaeological objects. However, it is very difficult to measure complicated objects, because of a relatively low-resolution. On the other hand, texture mapping can be a solution to complement the low resolution, and to generate 3D model with higher fidelity. But, a huge cost is required for the construction of textured 3D model, because huge labor is demanded, and the work depends on editor's experiences and skills . Moreover, the accuracy of data would be lost during the editing works. In this research, using the laser profiler and a non-calibrated digital camera, a method is proposed for the automatic generation of 3D model by integrating these data. At first, region segmentation is applied to laser range data to extract geometric features of an object in the laser range data. Various information such as normal vectors of planes, distances from a sensor and a sun-direction are used in this processing. Next, an image segmentation is also applied to the digital camera images, which include the same object. Then, geometrical relations are determined by corresponding the features extracted in the laser range data and digital camera’ images. By projecting digital camera image onto the surface data reconstructed from laser range image, the 3D texture model was generated automatically.

• 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.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1404-1407
• /
• 2002

Although we look at objects as 2D images through our eyes, we can reconstruct the shape and/or depth of objects. In order to realize this ability using computers, it is required that the method which can estimate the 3D features of object from 2D images. As feature which represents 3D shapes effectively, three dimensional vector autoregressive model is pro- posed. If this feature is associated other feature of 2D shape, then above aim might be achieved. On the other hand, as feature which represents 2D shapes, quasi moment features is proposed. As the first step of association of these features, we constructed real time simulator that computes both of two features concurrently from object data (3D curves) . This simulator can also rotate object and estimate the rotation The method using 3D VAR model estimates the rotation correctly, but the estimation by quasi moment features includes much errors. This reason would be that projected images are constructed by the points only, and doesn't have enough sizes to estimate the correct 3D rotation parameters.

• Journal of Information Processing Systems
• /
• 제15권5호
• /
• pp.1179-1191
• /
• 2019

We design an ingenious view-pooling method named learning-based multiple pooling fusion (LMPF), and apply it to multi-view convolutional neural network (MVCNN) for 3D model classification or retrieval. By this means, multi-view feature maps projected from a 3D model can be compiled as a simple and effective feature descriptor. The LMPF method fuses the max pooling method and the mean pooling method by learning a set of optimal weights. Compared with the hand-crafted approaches such as max pooling and mean pooling, the LMPF method can decrease the information loss effectively because of its "learning" ability. Experiments on ModelNet40 dataset and McGill dataset are presented and the results verify that LMPF can outperform those previous methods to a great extent.

• 한국운동역학회지
• /
• 제31권3호
• /
• pp.155-161
• /
• 2021

Objective: There is a lack of studies using the 3D-2D image registration techniques on the mechanism of a shoulder injury for ice hockey players. This study aimed to analyze in vivo 3D glenohumeral joint arthrokinematics in collegiate ice hockey athletes and compare shoulder scaption with or without a hockey stick using the 3D-2D image registration technique. Method: We recruited 12 male elite ice hockey players (age, 19.88 ± 0.65 years). For arthrokinematic analysis of the common shoulder abduction movements of the injury pathogenesis of ice hockey players, participants abducted their dominant arm along the scapular plane and then grabbed a stick using the same motion under C-arm fluoroscopy with 16 frames per second. Computed tomography (CT) scans of the shoulder complex were obtained with a 0.6-mm slice pitch. Data from the humerus translation distances, scapula upward rotation, anterior-posterior tilt, internal to external rotation angles, and scapulohumeral rhythm (SHR) ratio on glenohumeral (GH) joint kinematics were outputted using a MATLAB customized code. Results: The humeral translation in the stick hand compared to the bare hand moved more anterior and more superior until the abduction angle reached 40°. When the GH joint in the stick hand was at the maximal abduction of the scapula, the scapula was externally rotated 2~5° relative to 0°. The SHR ratio relative to the abduction along the scapular plane at 40° indicated a statistically significant difference between the two groups (p < 0.05). Conclusion: With arm loading with the stick, the humeral and scapular kinematics showed a significant correlation in the initial section of the SHR. Although these correlations might be difficult in clinical settings, ice hockey athletes can lead to the movement difference of the scapulohumeral joints with inherent instability.