• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 하계종합학술대회 논문집(4)
• /
• pp.222-225
• /
• 2000

Human frequently communicate non-linguistic information with gesture. So, we must develop efficient and fast gesture recognition algorithms for more natural human-computer interaction. However, it is difficult to recognize gesture automatically because human's body is three dimensional object with very complex structure. In this paper, we suggest a method which is able to detect key frames and frame changes, and to classify image sequence into some gesture groups. Gesture is classifiable according to moving part of body. First, we detect some frames that motion areas are changed abruptly and save those frames as key frames, and then use the frames to classify sequences. We symbolize each image of classified sequence using Principal Component Analysis(PCA) and clustering algorithm since it is better to use fewer components for representation of gestures. Symbols are used as the input symbols for the Hidden Markov Model(HMM) and recognized as a gesture with probability calculation.

• 제어로봇시스템학회논문지
• /
• 제10권3호
• /
• pp.267-272
• /
• 2004

In this Paper, we present analysis on the dynamics of human swing and its realization by a miniaturized humanoid robot. Since the motion of legs is the most important in the swing, the swing system can be approximated as a double pendulum. Based on Lagrangian analysis, the leg motion is designed to make the swing motion as sustained oscillation. In order to detect the peak instant of the swing and to synchronize the leg motion with the swing, we use ADXL acceleration/inclination sensor. The miniaturized humanoid in this paper has total 20 DOFs including 6 DOFs in each leg, 34cm in height, and 2kg in weight. As a result of realization of the swing by the humanoid, the sustained oscillation is verified through experiments.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2011년도 추계학술대회
• /
• pp.157-160
• /
• 2011

본 논문에서는 실시간 감시 시스템에서 녹화 도중에 검출된 움직임 벡터를 이용하여 사람의 활동성을 인식하고 분석하고자 한다. 전경에서 블랍(사람)을 검출하는 방법은 기존에 연구했던 차 영상을 이용하였고, MPEG-4 동영상 녹화 시 EPZS(Enhanced Predictive Zonal Search)에서 검출되는 움직임 벡터의 값을 이용하였다. 본 논문에서는 사람의 행동을 크게 세 가지의 {Active, Inactive}, {Moving, Non-moving}, {Walking, Running} 메타 클래스로 분류하고 인식하였다. 각 단계에서는 단계별 임계값을 이용하여 구분하였다. 실험을 위해서 약 150개의 상황을 연출하였으며, 실험 영상에서 각 단계를 구분하는데 약 86% ~ 98% 까지의 높은 인식률을 보였다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권12호
• /
• pp.1075-1082
• /
• 2007

본 논문은 동영상에 등장하는 다수 사람의 동작을 검출하여 검출된 동작을 개별적으로 인식하는 방법을 제안한다. 동작이 수행되는 속도 또는 크기 변화에 강인한 인식 성능을 갖기 위해 시공간축 피라미드(Spatial-Temporal Pyramid)방식을 적용한다. 동작 표현 방식을 통계적 특성 기반의 모션 그래디언트 히스토그램(MGH:Motion Gradient Histogram)으로 선택하여 인식 과정에서 발생하는 복잡도를 최소화 하였다. 다수의 동작을 검출하기 위하여 이진 차영상을 축적한 모션 에너지 이미지(MEI: Motion Energy Image) 방법을 적용하여 효율적으로 개별적 동작 영역을 획득한다. 각 영역은 동작 표현 방법인 MGH로 나타내어지고, 크기 변화에 강인하도록 피라미드 방식을 적응하여 학습된 템플릿 MGH와 유사도를 상호 비교하여 최종 인식 결과를 얻는다. 인식 성능의 평가를 위해 10개의 동영상을 활용하여 단일 객체, 다수 객체, 속도 및 크기 변화, 기존 방식과의 비교, 기타 추가 실험 등을 실시하여 다양한 조건의 영상에서 양호한 인식 결과를 확인 할 수 있었다.

• 대한의용생체공학회:의공학회지
• /
• 제16권2호
• /
• pp.217-222
• /
• 1995

We propose some applications of image processing techniques to extract quantitative measurements by using a camera system developed in Korea university and Catholic Medical School. From now on the system will be called as KCMOTION. The purpose of this study is to provide basic kinematic and kinetic data for the analysis of human movements and to find the clinical usefulness and reliability of the proposed motion analysis system. Two tests, sit-to-stand (STS) movements and pendulum test, are conducted by the system. The aims of the tests are to identify variability and reliability of KCMOTION to give some quantitative comparisons to the other systems. The result of STS movement are compared to the LOCUS IIID motion analyzer by the ratio of maximum flexion movement per body weight to the actual maximum flexion extension torque per body weight. That result in 29 % and 33 % for hip and knee joint, respectively in KCMOTION and 27 % and 30 % in LOCUS IIID System. The results of the pendulum movements are compared to that of using Cybex and Electrogoniometer with relaxation index, amplitude ratio, swing number and swing time. The results of relaxation index and amplitude ratio of the KCMOTION are between those of the Cybex and Electrogoniometer. We also observed that the KCMOTION detect more natural movement, from the results of swing number and time.

• 산업공학
• /
• 제14권1호
• /
• pp.95-105
• /
• 2001

Previous work measurement methods need much time and effort of time study analysts because they have to measure required time through direct observations. In this study, we propose a method which efficiently measures standard times without involvement of human analysts using digital image processing techniques. This method consists of two main steps: motion representation step and cycle segmentation step. In motion representation step, we first detect the motion of any object distinct from its background by differencing two consecutive images separated by a constant time interval. The images thus obtained then pass through an edge detector filter. Finally, the mean values of coordinates of significant pixels of the edge image are obtained. Through these processes, the motions of the observed worker are represented by two time series data of worker location in horizontal and vertical axes. In the second step, called the cycle segmentation step, we extract the frames which have maximum or minimum coordinates in one cycle and store them in a stack, and calculate each cycle time using these frames. In this step we also consider methods on how to detect work delays due to unexpected events such as operator's escapement from the work area, or interruptions. To condude, the experimental results show that the proposed method is very cost-effective and useful for measuring time standards for various work environment.

• 한국정밀공학회지
• /
• 제24권6호
• /
• pp.66-71
• /
• 2007

Development of a locomotive mechanism for the capsule type endoscopes will largely enhance their ability to diagnose disease of digestive organs. As a part of it, there should be provided a detection device of their position in human organs for the purpose of observation and motion control. In this paper, a permanent magnet outside human body was employed to project magnetic field on a capsule type endoscope, while its position dependent flux density was measured by three hall-effect sensors which were orthogonally installed inside the capsule. In order to detect the 2-D position data of the capsule with three hall-effect sensors including the roll, pitch and yaw angle, the permanent magnet was extra translated during the measurement. In this way, the 2-D coordinates and three rotation angles of a capsule endoscope on the same motion plane with the permanent magnet could be detected. The working principle and performance test results of the capsule position detection device were introduced in this paper showing that they could be also applied to 6-DOF position detection.

• 한국멀티미디어학회논문지
• /
• 제25권4호
• /
• pp.575-582
• /
• 2022

In the manufacturing environment, research to minimize robot collisions with human beings have been widespread, but in order to interact with robots, it is important to precisely recognize and predict human actions. In this research, after enhancing performance by applying group normalization to the Hourglass model to detect the operator motion, the skeleton was estimated and data were created using this model. And then, three types of operator's movements were recognized using LSTM. As results of the experiment, the accuracy was enhanced by 1% using group normalization, and the recognition accuracy was 99.6%.

• 전기학회논문지
• /
• 제67권7호
• /
• pp.903-909
• /
• 2018

We propose laser scanner sensor system based walking pattern and tracking method of multiple human. This system uses laser scanners sensors and is applicable to wide and crowded area such as hospital and medical care center. The primary objective of this research is to promote the development of robust, repeatable and transferable software for security system that can automatically detect, track and follow people in public area. We developed the method of human identification for this system. Our method is following: 1. Best-walking pattern data are obtained by the help of human position and direction data obtained by laser scanners. 2. Human identification is conducted by calculating the correlation between the step length of walking human. It becomes possible to conduct human identification even in crowded scenes by estimating the movements of waling human' feet are periodic. In the experiment in the station, some effectiveness of this method became clear.

• 스마트미디어저널
• /
• 제5권3호
• /
• pp.49-53
• /
• 2016

인체 전자기장 신호를 검출하여 동작 인식에 이용하는 비접촉방식의 방법은 시간과 공간의 제약이 기존의 시스템보다 덜하므로 관련 연구들이 진행 중에 있다. 본 논문에서는 비접촉방식의 인체전기장 신호를 검출할 수 있는 하드웨어를 구현하여 이를 디지털 파형화 하여 인식률을 높일 수 있는 회로시스템을 설계하였다. 차동 증폭회로의 구현과 비교기를 연동한 디지털 파형화를 위한 회로 시스템을 시뮬레이션과 결합하여 PCB화한 후/ 설계된 전체 회로 시스템에 대한 특성평가를 수행하였다.