• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.307-312
• /
• 2001

In this paper, we proposed the object-tracking algorithm that recognizes and estimates the any shaped and size objects using vision system. For the extraction of the object from the background of the acquired images, spatio-temporal filter and signature parsing algorithm are used. Specially, for the solution of correspondence problem of the multiple objects tracking, we compute snake energy and position information of the target objects. Through the real-time tracking experiment, we verified the effectiveness of the suggested tracking algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1202-1205
• /
• 2005

This paper describes a method for vision-based person identification that can detect, track, and recognize person from video using multiple cues: height and dressing colors. The method does not require constrained target's pose or fully frontal face image to identify the person. First, the system, which is connected to a pan-tilt-zoom camera, detects target using motion detection and human cardboard model. The system keeps tracking the moving target while it is trying to identify whether it is a human and identify who it is among the registered persons in the database. To segment the moving target from the background scene, we employ a version of background subtraction technique and some spatial filtering. Once the target is segmented, we then align the target with the generic human cardboard model to verify whether the detected target is a human. If the target is identified as a human, the card board model is also used to segment the body parts to obtain some salient features such as head, torso, and legs. The whole body silhouette is also analyzed to obtain the target's shape information such as height and slimness. We then use these multiple cues (at present, we uses shirt color, trousers color, and body height) to recognize the target using a supervised self-organization process. We preliminary tested the system on a set of 5 subjects with multiple clothes. The recognition rate is 100% if the person is wearing the clothes that were learned before. In case a person wears new dresses the system fail to identify. This means height is not enough to classify persons. We plan to extend the work by adding more cues such as skin color, and face recognition by utilizing the zoom capability of the camera to obtain high resolution view of face; then, evaluate the system with more subjects.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.8
• /
• pp.1802-1808
• /
• 2003

In this paper, we propose a method about the extraction of the pedestrian tracking trajectory in the road and we used the method of mosaic background extraction and adaptive neural network for automatic pedestrian tracking system. We used mosaic background extraction to overcome ghost phenomenon. And we detected pedestrian using differential image analysis. We used adaptive neural network for multiple pedestrian tracking that non­rigid form moving. The ART2 network is capable of detecting the mass­centers of moving objects within one frame. The history of neurons positions in the sequential frames approximates the traces of the targets. The experiments done with the network in simulated environment show promising results.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.68-73
• /
• 2005

Various techniques have been proposed for detection and tracking of targets in order to develop a real-world computer vision system, e.g., visual surveillance systems, intelligent transport systems (ITSs), and so forth. Especially, the idea of distributed vision system is required to realize these techniques in a wide-spread area. In this paper, we develop a ubiquitous vision system for location-awareness of multiple targets. Here, each vision sensor that the system is composed of can perform exact segmentation for a target by color and motion information, and visual tracking for multiple targets in real-time. We construct the ubiquitous vision system as the multiagent system by regarding each vision sensor as the agent (the vision agent). Therefore, we solve matching problem for the identity of a target as handover by protocol-based approach. We propose the identified contract net (ICN) protocol for the approach. The ICN protocol not only is independent of the number of vision agents but also doesn't need calibration between vision agents. Therefore, the ICN protocol raises speed, scalability, and modularity of the system. We adapt the ICN protocol in our ubiquitous vision system that we construct in order to make an experiment. Our ubiquitous vision system shows us reliable results and the ICN protocol is successfully operated through several experiments.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.8
• /
• pp.1476-1483
• /
• 2007

In a multi-sensor target tracking systems, the local sensors have the role of tracking the target and transferring the measurements to the fusion center. The measurements from the same target can arrive out of sequence called, the out-of-sequence measurements(OOSMs). The OOSM can arise in a form of single-lag or multi-lag throughout the transfer at the fusion center. The recursive retrodiction step was proposed to update the current state estimates with the multi-lag OOSM from the several previous papers. The real world has the possible situations that the maneuvering target informations can arrive at the fusion center with the random clutter in the possible OOSMs. In this paper, we incorporate the IMM-MPDA(Interacting Multiple Model - Most Probable Data Association) into the multi-lag OOSM update. The performance of the IMM-MPDA filter with multi-lag OOSM update is analyzed for the various clutter densities, OOSM lag numbers, and target maneuvering indexes. Simulation results show that IMM-MPDA is sufficient to be used in out of sequence environment and it is necessary to correct the current state estimates with OOSM except a very old OOSM.

• Journal of Intelligence and Information Systems
• /
• v.15 no.1
• /
• pp.51-64
• /
• 2009

A location tracking sensor such as GPS, INS, Radar, and optical equipments is used in tracking Maneuvering Targets with a multi-sensor, and such systems are used to track, detect, and control UAV, guided missile, and spaceship. Until now, Most of the studies related to tracking Maneuvering Targets are on fusing multiple Radars, or adding a supplementary sensor to INS and GPS. However, A study is required to change the degree of application in fusions since the system property and error property are different from sensors. In this paper, we perform the error analysis of the sensor properties by adding a ground radar to GPS and INS for improving the tracking performance by multi-sensor fusion, and suggest the tracking algorithm that improves the precision and stability by changing the sensor probability of each sensor according to the error. For evaluation, we extract the altitude values in a simulation for the trajectory of UAV and apply the suggested algorithm to carry out the performance analysis. In this study, we change the weight of the evaluated values according to the degree of error between the navigation information of each sensor to improve the precision of navigation information, and made it possible to have a strong tracking which is not affected by external purposed environmental change and disturbance.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.13 no.4
• /
• pp.298-306
• /
• 2013

Recent years have witnessed a growing interest in the fields of video surveillance and mobile object tracking. This paper proposes a mobile object tracking algorithm. First, several parameters such as object window, object area, and expansion-contraction (E-C) parameter are defined. Then, a modified E-C algorithm for multiple-object tracking is presented. The proposed algorithm tracks moving objects by expansion and contraction of the object window. In addition, it includes methods for updating the background image and avoiding occlusion of the target image. The validity of the proposed algorithm is verified experimentally. For example, the first scenario traces the path of two people walking in opposite directions in a hallway, whereas the second one is conducted to track three people in a group of four walkers.

• Journal of Advanced Navigation Technology
• /
• v.21 no.1
• /
• pp.119-125
• /
• 2017

In this paper, we propose a method that merges and splits players in the MeanShift tracking framework. The MeanShift tracking moves the center of tracking window to the maximum probability location given the target probability distribution. This tracking method has been widely used for real-time tracking problems because of its fast processing speed. However, it hardly handles occlusions in multiple object tracking systems. Occlusions can be usually solved by applying data association methods. In this paper, we propose a method that can be applied before data association methods. The proposed method automatically merges and splits the overlapped players by adjusting the each player's tracking map. We have compared the tracking performance of the MeanSfhit tracking algorithm and the proposed method.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.15 no.3
• /
• pp.343-349
• /
• 2012

To detect targets for autonomous navigation of unmanned ground vehicle, mounted sensors are required to work all-weather condition. In this point of view, the FMCW radar is quietly appropriate. In this paper, we present development results of target signal simulator for multi-beam type FMCW radar. A target signal simulator make pseudo target signals which simulates multiple moving targets. And we describe how to make hit information for each target in multi-beam type radar. The developed methods are utilized for target tracking device. Moreover it can be applied to similar target signal simulator.

• Journal of Advanced Navigation Technology
• /
• v.4 no.1
• /
• pp.50-66
• /
• 2000

For a long time target maneuvers in tracking problem have been a difficult task to handle. Once a maneuvering such as abrupt change in target accelerations occur, the tracking fiter no longer yields a reasonable estimate of the target position. In order to resolve this cumbersome maneuvering problem. Advanced methods have here proposed : Colored noise, IE(Input Estimation), VD(Variable Dimension), IMM(Interaction Multiple Model), Jump-type processes and jerk model, etc. In this paper, tracking performance of the jerk model is analyzed. Jerk model in which the derivative of target acceleration is included as a state recently attracted considerable attraction. Firstly 3-dimensional Kalman filter is described on the basis of jerk model. Then using this filter, Monte-Carlo simulations are carried out and the filter formance with respect to the variation of jerk time-constant is analyzed. Especially, since jerk model's transient performance is expected to be poor, the performance of analysis of transient response of the model is included too.