• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.359-362
• /
• 2001

본 논문에서는 광학센서와 적외선 센서를 사용하는 Multi-sensor 시스템에서 영상 정보를 통한 물체의 추적 및 인식에 필요한 영상을 분리하는데 필요한 전처리와 object 기반의 추적 방법을 제안하였다. 일반적인 추적 알고리즘의 목표는 consistency를 유지하는데 있다. 그러나 인식에 필요한 영상을 분리하기 위해서는 물체의 범위를 정확히 판단 할 수 있는 능력이 중요하다. 이를 위해 CCD와 IR영상에 동시에 적용 가능한 전처리 기법과 object 기반의 two-step 추적 알고리즘을 통해 consistency외에도, 물체의 범위를 estimation하여 인식에 필요한 범위를 분리해 낸다. 본 논문에서는 ITS 의 ETCS application을 위해 이종 센서인 CCD와 IR의 야간 차량 영상정보를 이용하여 알고리즘을 test 하였다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.4
• /
• pp.38-45
• /
• 2001

In this paper, we propose an algorithm for moving object tracking from moving camera using histogram back program intersection(HI) and XY-projection The proposed method segments objects using histogram back projection, matches tracing objects using histogram intersection and extracts them using XY- projection. Through the simulation this paper shows that the proposed method segments. matches and tracks objects without significant error image sequences obtained by moving camera.

• Proceedings of the KIEE Conference
• /
• 2008.04a
• /
• pp.113-114
• /
• 2008

본 논문에서는 HD 비디오 영상 처리를 효과적으로 수행할 수 있는 다중 DSP 아키텍쳐를 제안하고 프로토타입 보드를 설계 제작하였다. 또한, 구현된 보드를 이용하여 비디오 영상 내 물체(얼굴)의 실시간 추석시스템을 구현하였다. 물체 추적 기법인 PF(Particle Filtering) 기법은 배경 클러터가 존재하는 환경에서도 강인하게 물체를 추적할 수 있지만 많은 수의 샘플을 사용하는 경우 필요한 계산량이 많아져 실시간 구현이 매우 어렵다는 문제점을 가지고 있다. 본 논문에서는 이러한 경우에도 실시간 추적이 가능하도록 병렬화된 PF 추적 방법을 제안하고 제작된 보드 상에 구현하였다. 구현된 병렬 처리 추적에서는 150개의 PF 샘플들을 5개의 슬레이브 DSP로 분산하여 컬러 유사도 기탄의 관측 확률을 계산하고 그 결과를 마스터 DSP에서 종합하여 추적의 정확도를 높이고자 하였다. 실험에는 $720{\times}480$ 픽셀 영상이 사용되었으며, 실험 결과 배경 클러터가 존재하는 경우에도 충분한 PF 샘플 수의 사용에 따라 대상 물체를 강인하게 추적하는 우수한 성능을 확인할 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.521-525
• /
• 2006

본 논문에서는 초음파 및 적외선 센서와 무선 카메라를 장착한 소형 이동 로봇의 장애물 회피 및 물체 추적 방법을 제시한다. 장애물 회피를 위해서 제어부의 초음파 발생 신호의 귀환시간과 거리와의 관계 및 적외선 센서에서 측정한 아날로그신호와 거리와의 관계를 추출하여 이동 로봇과 물체와의 거리를 판단하여 로봇의 움직임을 제어하는데 사용한다. 물체 추적 모드에서는 첫째, 물체와 배경 및 유사잡음들과의 강인한 분리를 위하여 고유색상정보와 움직임 정보 등의 사전정보를 활용하였으며 둘째, 형태의 변화가 수반되는 경우에도 유연한 대처능력을 갖도록 하기 위해 영상의 영역분할 방법을 통해 모든 후보영역내의 물체의 존재를 확인하고 물체영역만을 추출하였다. 셋째, 물체 형태정보함수를 정의하고 해당함수를 형태의 보전 에너지로 활용하여 동일 물체의 대응문제를 효과적으로 해결하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.1
• /
• pp.35-41
• /
• 2015

When a video surveillance system tracks a specific object, it is very important to get quickly the information of the object through fast image processing. Usually one camera surveillance system for tracking the object made results in various problems such like occlusion, image noise during the tracking process. It makes difficulties on image based moving object tracking. Therefore, to overcome the difficulties the multi video surveillance system which installed several camera within interested area and looking the same object from multi angles of view could be considered as a solution. If multi cameras are used for tracking object, it is capable of making a decision having high accuracy in more wide space. This paper proposes a method of recognizing and tracking a specific object like a car using the homography in which multi cameras are installed at the crossroad.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2001.05a
• /
• pp.633-637
• /
• 2001

In this paper, we proposed model creation algorithm for multiple moving objects tracking. The proposed algorithm is divided that the initial model creation step as moving objects are entered into background image and the model reformation step in the moving objects tracking step. In the initial model creation step, the initial model is created by AND operating division image, divided using difference image and clustering method, and edge image of the current image. In the model reformation step, a new model was reformed in the every frame to adapt appearance change of moving objects using Hausdorff Distance and 2D-Logarithmic searching algorithm. We simulated for driving cart in the road. In the result, model was created over 98% in case of irregular approach direction of cars and tracking objects number.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.6
• /
• pp.743-748
• /
• 2011

In this paper, we propose a collaborative tracking algorithm for intelligent video surveillance systems using the multiple network cameras. To do this, each camera detects a moving object and it's movement direction by motion templates. Once a moving object is detect, the Kalman filter is used to reduce noises, and a collaborative tracking camera is selected according to the movement direction and the camera state. In this procedure, Pan-Tilt-Zoom(PTZ) parameters are assigned to obtain clear images. Finally, some experiments show the validity of the proposed method.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.2
• /
• pp.28-34
• /
• 1998

In this paper, I'd like to make a suggestion for boundary line detect algorithm which is used 3-D image processing system in order to track moving object. Through this study, more than anything else, difference image method was adopted to detect moving object in input image. To detect moving object, I made use of detect windows constructed by 4's predictive areas and object area for the purpose of reducing processing time and its size was determined by the size of moving object and prediction parameter directed center position. And also, tracking camera was movable toward the direction of X, Y by DC motor. As a conclusion of the study proposed algorithm, I found out the following results that tracking error was less than 6% of total moving object size and maximum tracking time 2 seconds by toy-car simulation.

• 전자공학회논문지 IE
• /
• v.43 no.2
• /
• pp.54-62
• /
• 2006

In this paper, a new stereo object tracking system using the disparity motion vector is proposed. In the proposed method, the time-sequential disparity motion vector can be estimated from the disparity vectors which are extracted from the sequence of the stereo input image pair and then using these disparity motion vectors, the area where the target object is located and its location coordinate are detected from the input stereo image. Basing on this location data of the target object, the pan/tilt embedded in the stereo camera system can be controlled and as a result, stereo tracking of the target object can be possible. From some experiments with the 2 frames of the stereo image pairs having $256\times256$ pixels, it is shown that the proposed stereo tracking system can adaptively track the target object with a low error ratio of about 3.05 % on average between the detected and actual location coordinates of the target object.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.2C
• /
• pp.166-174
• /
• 2006

In this paper, a new stereo object tracking system using the disparity motion vector is proposed. In the proposed method, the time-sequential disparity motion vector can be estimated from the disparity vectors which are extracted from the sequence of the stereo input image pair and then using these disparity motion vectors, the area where the target object is located and its location coordinate are detected from the input stereo image. Being based on this location data of the target object, the pan/tilt embedded in the stereo camera system can be controlled and as a result, stereo tracking of the target object can be possible. From some experiments with the 2 frames of the stereo image pairs having 256$\times$256 pixels, it is shown that the proposed stereo tracking system can adaptively track the target object with a low error ratio of about 3.05$\%$ on average between the detected and actual location coordinates of the target object.