• 한국멀티미디어학회논문지
• /
• 제12권4호
• /
• pp.521-529
• /
• 2009

본 논문은 비전 기반 동작 인식 방법으로 모범 동작의 유형을 모형화하고 이를 이용하여 사용자의 동작을 인식하고 모범동작과 사용자의 동작간의 유사도를 측정하는 방법을 제안한다. 동작 인식을 위하여 은닉 마르코프 모델 기반의 유형화 기법을 통하여 모범 동작의 유형 모델을 구성하고 이를 이용하여 사용자의 동작을 인식한다. 유사도 측정을 위하여 편집 거리 알고리즘을 응용하여 모범 동작과 사용자 동작의 유사도를 측정하고 점수 표기가 가능하도록 하였다. 본 논문에서 제안하는 동작 인식 처리 방법은 평균 93% 이상의 높은 인식율을 보였다. 본 연구의 결과는 동작 인식 기반 게임, 자세인식, 동작의 반복 훈련 및 훈련 달성도 측정을 요하는 재활훈련 시스템 등에 활용 가능하다.

• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.1109-1116
• /
• 2020

사람이 따라 하는 능력을 평가하는 스코어는 아동의 발달 단계 혹은 골프, 무용 동작 등을 점검하는 좋은 수단이 될 수 있다. 또한, 이는 AR, VR 응용에서 HCI로도 활용될 수 있다. 본 논문에서는 동작을 주도해서 수행하는 시범자와 그 동작을 따라 하는 참여자 간의 동작 유사도를 평가하는 방안을 제시하고, 여기서 우리는 Openpose의 키포인트 벡터 유사도의 유클리디안 L2 거리를 활용하는 동작 유사도를 제안한다. 제안된 기법은 DTW를 사용하기 때문에 시간 지연차가 있는 동작에 유연하게 대처할 수 있다.

• 방송공학회논문지
• /
• 제12권3호
• /
• pp.289-292
• /
• 2007

A background initialization algorithm based on the spatiotemporal similarity measure in a motion tracking system is proposed. From the accumulated difference between the base frame and the other frames in a subinterval, the regions affected by moving objects are located. The median is applied over the subsequence in the subinterval in which co-located regions share the similarity. The outputs from each subinterval are filtered by second stage median filter. The proposed method showed good results even in the busy and crowded sequences where the real background does not exit.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.903-906
• /
• 1998

We propose a new method for merging partical regions which have similar motion characteristics. In the proposed method, the trajectories of each partial regions are computed and the regions consisting a same object are merged by a similarity function defined in this paper. The similarity function is defined to take into consideration the motion characteristics of the trajectories. The proposed method shows good performance when the regions of an object show little variation in size and shape.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제6권2호
• /
• pp.333-346
• /
• 2014

If we want to analyze the motion of a body in fluid, we should use rigid-body dynamics and fluid dynamics together. Even if the rigid-body and fluid dynamics are each self-consistent, there arises the problem of self-similar structure in the equation of motion when the two dynamics are coupled with each other. When the added mass is greater than the mass of a body, the calculated motion is divergent because of its self-similar structure. This study showed that the above problem is an inherent problem. This problem of self-similar structure may arise in the equation of motion in which the fluid dynamic forces are treated as external forces on the right hand side of the equation. A reconfiguration technique for the equation of motion using pseudo-added-mass was proposed to resolve the self-similar structure problem; specifically for the case when the fluid force is expressed by integration of the fluid pressure.

• 대한수학회보
• /
• 제52권6호
• /
• pp.2071-2093
• /
• 2015

In this paper, we investigate the similarity transformations in the Minkowski n-space. We study the geometric invariants of non-null curves under the similarity transformations. Besides, we extend the fundamental theorem for a non-null curve according to a similarity motion of ${\mathbb{E}}_1^n$. We determine the parametrizations of non-null self-similar curves in ${\mathbb{E}}_1^n$.

• 한국컴퓨터정보학회논문지
• /
• 제28권1호
• /
• pp.39-47
• /
• 2023

본 논문에서는 골프 동영상 속 스윙 자세 사이의 유사도를 측정할 수 있는 방법을 제안한다. 딥러닝 기반 인공지능 기술이 컴퓨터 비전 분야에 효과적인 것이 알려지면서 동영상을 기반으로 한 스포츠 데이터 분석에 인공지능을 활용하기 위한 시도가 증가하고 있다. 본 연구에서는 딥러닝 기반의 자세 추정 모델을 사용하여 골프 스윙 동영상 속 사람의 관절 좌표를 획득하였고, 이를 바탕으로 각 스윙 구간별 유사도를 측정하였다. 제안한 방법의 평가를 위해 GolfDB 데이터셋의 Driver 스윙 동영상을 활용하였다. 총 36명의 선수에 대해 스윙 동영상들을 두 개씩 짝지어 스윙 유사도를 측정한 결과, 본인의 또 다른 스윙이 가장 유사하다고 평가한 경우가 26명이었으며, 이때의 유사도 평균 순위는 약 5위로 확인되었다. 이로부터 비슷한 동작을 수행하고 있는 경우에도 면밀히 유사도를 측정하는 것이 가능함을 확인할 수 있었다.

• International journal of advanced smart convergence
• /
• 제8권2호
• /
• pp.116-125
• /
• 2019

The fire should extinguish as soon as possible because it causes economic loss and loses precious life. In this study, we propose a new atypical fire and smoke detection algorithm using deep learning and color histogram of fire and smoke. First, input frame images obtain from the ONVIF surveillance camera mounted in factory search motion candidate frame by motion detection algorithm and mean square error (MSE). Second deep learning (Faster R-CNN) is used to extract the fire and smoke candidate area of motion frame. Third, we apply a novel algorithm to detect the fire and smoke using color histogram algorithm with local area motion, similarity, and MSE. In this study, we developed a novel fire and smoke detection algorithm applied the local motion and color histogram method. Experimental results show that the surveillance camera with the proposed algorithm showed good fire and smoke detection results with very few false positives.

• 한국통신학회논문지
• /
• 제22권6호
• /
• pp.1195-1209
• /
• 1997

This paper presents a new morphological spatio-temporal segmentation algorithm. The algorithm incorporates luminance and motion information simultaneously, and uses morphological tools such as morphological filtersand watershed algorithm. The procedure toward complete segmentation consists of three steps:joint marker extraction, boundary decision, and motion-based region fusion. First, the joint marker extraction identifies the presence of homogeneours regions in both motion and luminance, where a simple joint marker extraction technique is proposed. Second, the spatio-temporal boundaries are decided by the watershed algorithm. For this purposek, a new joint similarity measure is proposed. Finally, an elimination ofredundant regions is done using motion-based region function. By incorporating spatial and temporal information simultaneously, we can obtain visually meaningful segmentation results. Simulation results demonstratesthe efficiency of the proposed method.

• 한국전문물리치료학회지
• /
• 제13권1호
• /
• pp.54-60
• /
• 2006

This study aimed to compare movement patterns of shoulder joints between the right and left symmetry in stroke patients and control subjects. This study proposes use of the voluntary response index (VRI) calculated from quantitative analysis of surface electromyographic (sEMG) and motion data recorded during voluntary movement as a feeding task. The VRI is comprised of two numeric values, one derived from the total muscle activity recorded for the voluntary motor task (magnitude), and the other from the sEMG distribution across the recorded muscles with the similarity index (SI). Five stroke patients and five age-matched healthy controls were recruited. Feeding motion was performed using the provided spoon five times with rests taken on a chair in between tasks. EMG data were digitized and analyzed on the basis of the root mean square (RMS) envelope of activity. The average amplitude of responses was calculated. Responsiveness and clinically meaningful levels of discrimination between stroke patients and control for EMG magnitude and SI were determined. The similarity index of the results from two successive examinations of both sides apart for stroke patients and control subjects were .86 and .95 in motion analysis and .84 and .99 in electromyographic analysis. The SI of sEMG data and motion data was significantly correlated in stroke patients. The data suggest that SI is a sensitive program for comparing and analyzing the symmetry of muscle activity and motion in both sides. This analysis method has a clinical value in grading muscular activity and movement impairment after brain injury.