• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.498-500
• /
• 2012

비디오 해석 및 추적기술은 특정한 시스템에서만 적용되는 것이 아니다. 이것은 비디오 내에서 의미 있는 정보를 능동적으로 감시 대상을 정의, 해석, 모델화, 추정 및 추적 할 수 있는 기반 기술을 의미하다. 일반적으로 감시시스템에서 감시 대상은 사람이나 차량이며, 상황에 따라 출입통제 구역으로 설정하기도 한다. 이는 연속된 영상에서 객체의 형태, 모양, 행동 분석, 움직임, 색상정보를 가지고 데이터 정의, 검출, 모델화를 통하여 인식, 식별 그리고 추적한다. 본 논문에서는 비디오 영상분석을 통해 단일카메라기반의 감시시스템과 PTZ 카메라기반 감시시스템 제안한다. 이때 단일 카메라기반의 감시는 배경생성방법을 이용하여 연속된 영상내의 객체를 지속적으로 관리가 가능하도록 설계하였고, PTZ 카메라기반의 감시는 카메라의 이동에 따른 배경안정화 방법과 카메라의 절대좌표를 활용하여 카메라 이동을 제어함과 동시에 오검출 문제를 해결하였다. 실험 및 결과분석으로는 시나리오 환경에서 배경생성방법을 이용한 검출의 정확성과 PTZ카메라 위치 변화에도 강인한 검출 결과를 비교 분석하였다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.5
• /
• pp.16-22
• /
• 2011

This paper presents the multi-object tracking approach using the background difference and particle filtering by monte carlo sampling. We apply particle filters based on probabilistic importance sampling to multi-object independently. We formulate the object observation model by the histogram distribution using color information and the object dynaminc model for the object motion information. Our approach does not increase computational complexity and derive stable performance. We implement the whole Bayesian maximum likelihood framework and describes robust methods coping with the real-world object tracking situation by the observation and transition model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.4
• /
• pp.851-859
• /
• 2011

In this paper, we propose the method that automatically extracts the silhouette and the joints of consecutive input image, and track joints to trace object for interaction between human and computer. Also the proposed method presents the action of human being to map human body using joints. To implement the algorithm, we model human body using 14 joints to refer to body size. The proposed method converts RGB color image acquired through a single camera to hue, saturation, value images and extracts body's silhouette using the difference between the background and input. Then we automatically extracts joints using the corner points of the extracted silhouette and the data of body's model. The motion of object is tracted by applying block-matching method to areas around joints among all image and the human's motion is mapped using positions of joints. The proposed method is applied to the test videos and the result shows that the proposed method automatically extracts joints and effectively maps human body by the detected joints. Also the human's action is aptly expressed to reflect locations of the joints

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.10
• /
• pp.43-53
• /
• 2012

In the this paper, a gesture-based user interface based on object detection using predefined gesture and the tracking of the detected object is proposed. For object detection, moving objects in a frame are computed by comparing multiple previous frames and predefined gesture is used to detect the target object among those moving objects. Any object with the predefined gesture can be used to control. We also propose an object tracking algorithm, namely density based meanshift algorithm, that uses color distribution of the target objects. The proposed object tracking algorithm tracks a target object crossing the background with a similar color more accurately than existing techniques. Experimental results show that the proposed object detection and tracking algorithms achieve higher detection capability with less computational complexity.

• Journal of Broadcast Engineering
• /
• v.7 no.2
• /
• pp.181-191
• /
• 2002

Video based surveillance systems based on an active camera require a fast algorithm for real time detection and tracking of local motion in the presence of global motion. This paper presents a new fast and efficient motion detection and tracking algorithm using the displaced frame difference (DFD). In the Proposed algorithm, first, a Previous frame is adaptively selected according to the magnitude of object motion, and the global motion is estimated by using only a few confident matching blocks for a fast and accurate result. Then, a DFD is obtained between the current frame and the selected previous frame displaced by the global motion. Finally, a moving object is extracted from the noisy DFD by utilizing the correlation between the DFD and current frame. We implement this algorithm into an active camera system including a pan-tilt unit and a standard PC equipped with an AMD 800MHz processor. The system can perform the exhaustive search for a search range of 120, and achieve the processing speed of about 50 frames/sec for video sequences of 320$\times$240. Thereby, it provides satisfactory tracking results.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.8
• /
• pp.1919-1925
• /
• 2013

This paper proposes a background image extraction algorithm for traffic environment analysis in a school zone. The proposed algorithm solves the problems by level changes and stationary objects to be occurred frequently in traffic environment. For the former, it renews rapidly the background image toward the current frame using a fast Sima-Delta algorithm and for the latter, it excludes the stationary objects from the background image by detecting dynamic regions using a just previous frame and a background image averaged for a long time. The results of experiments show that the proposed algorithm adapts quickly itself to level change well, and reduces about 40~80% of SAD in background region in comparison with the conventional algorithms.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.3
• /
• pp.61-69
• /
• 2010

This paper presents an efficient object selection algorithm by analyzing and detecting of human activity. Generally, when people point any something, they will put a face on the target direction. Therefore, the direction of the face and fingers and was ordered to be connected to a straight line. At first, in order to detect the moving objects from the input frames, we extract the interesting objects in real time using background subtraction. And the judgment of movement is determined by Principal Component Analysis and a designated time period. When user is motionless, we estimate the user's indication by estimation in relation to vector from the head to the hand. Through experiments using the multiple views, we confirm that the proposed algorithm can estimate the movement and indication of user more efficiently.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.3
• /
• pp.35-41
• /
• 2018

In this paper, we define a haptic rendering algorithm for a situation that has collision between static object and single object. We classified video scenes into four categories which can be easily seen in video sequence. The proposed algorithm can detect which frame is suitable for haptic rendering by detecting the change of direction using motion estimation and change of shape using object tracking. As a result, a total of 13 frames are extracted from the sample video and playing time of these frames were calculated. We confirmed that the haptic effect appears in expected playing time by adding the appropriate haptic generating waveform thtough the haptic editing program.

• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.114-125
• /
• 2002

In this paper, we are proposing a hierarchical segmentation method that first segments the video data into units of shots by detecting cut and dissolve, and then decides types of camera operations or object movements in each shot. In our previous work[1], each picture group is divided into one of the three detailed categories, Shot(in case of scene change), Move(in case of camera operation or object movement) and Static(in case of almost no change between images), by analysing DC(Direct Current) component of I(Intra) frame. In this process, we have designed two-stage hierarchical neural network with inputs of various multiple features combined. Then, the system detects the accurate shot position, types of camera operations or object movements by searching P(Predicted), B(Bi-directional) frames of the current picture group selectively and hierarchically. Also, the statistical distributions of macro block types in P or B frames are used for the accurate detection of cut position, and another neural network with inputs of macro block types and motion vectors method can reduce the processing time by using only DC coefficients of I frames without decoding and by searching P, B frames selectively and hierarchically. The proposed method classified the picture groups in the accuracy of 93.9-100.0% and the cuts in the accuracy of 96.1-100.0% with three different together is used to detect dissolve, types of camera operations and object movements. The proposed types of video data. Also, it classified the types of camera movements or object movements in the accuracy of 90.13% and 89.28% with two different types of video data.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.111-121
• /
• 2011

Intelligent video analysis systems require techniques which can predict accidents and provide alarms to the monitoring personnel. In this paper, we present an abnormal behavior analysis technique based on adaptive background generation. More specifically, abnormal behaviors include fence climbing, abandoned objects, fainting persons, and loitering persons. The proposed video analysis system consists of (i) background generation and (ii) abnormal behavior analysis modules. For robust background generation, the proposed system updates static regions by detecting motion changes at each frame. In addition, noise and shadow removal steps are also were added to improve the accuracy of the object detection. The abnormal behavior analysis module extracts object information, such as centroid, silhouette, size, and trajectory. As the result of the behavior analysis function objects' behavior is configured and analyzed based on the a priori specified scenarios, such as fence climbing, abandoning objects, fainting, and loitering. In the experimental results, the proposed system was able to detect the moving object and analyze the abnormal behavior in complex environments.