• 대한원격탐사학회지
• /
• 제40권1호
• /
• pp.115-122
• /
• 2024

Multi-object tracking (MOT) is a vital component in understanding the surrounding environments. Previous research has demonstrated that MOT can successfully detect and track surrounding objects. Nonetheless, inaccurate classification of the tracking objects remains a challenge that needs to be solved. When an object approaching from a distance is recognized, not only detection and tracking but also classification to determine the level of risk must be performed. However, considering the erroneous classification results obtained from the detection as the track class can lead to performance degradation problems. In this paper, we discuss the limitations of classification in tracking under the classification uncertainty of the detector. To address this problem, a class update module is proposed, which leverages the class uncertainty estimation of the detector to mitigate the classification error of the tracker. We evaluated our approach on the VisDrone-MOT2021 dataset,which includes multi-class and uncertain far-distance object tracking. We show that our method has low certainty at a distant object, and quickly classifies the class as the object approaches and the level of certainty increases.In this manner, our method outperforms previous approaches across different detectors. In particular, the You Only Look Once (YOLO)v8 detector shows a notable enhancement of 4.33 multi-object tracking accuracy (MOTA) in comparison to the previous state-of-the-art method. This intuitive insight improves MOT to track approaching objects from a distance and quickly classify them.

• IEIE Transactions on Smart Processing and Computing
• /
• 제2권1호
• /
• pp.27-35
• /
• 2013

This paper proposes a face tracking algorithm for a viewpoint adaptive multi-view synthesis system. The original scene captured by a depth camera contains a texture image and 8 bit gray-scale depth map. From this original image, multi-view images that correspond to the viewer's position can be synthesized using geometrical transformations, such as rotation and translation. The proposed face tracking technique gives a motion parallax cue by different viewpoints and view angles. In the proposed algorithm, the viewer's dominant face, which is established initially from a camera, can be tracked using the statistical characteristics of face colors and deformable templates. As a result, a motion parallax cue can be provided by detecting the viewer's dominant face area and tracking it, even under a heterogeneous background, and synthesized sequences can be displayed successfully.

• 한국추진공학회:학술대회논문집
• /
• 한국추진공학회 2006년도 제27회 추계학술대회논문집
• /
• pp.299-302
• /
• 2006

본 연구에서는 화학적으로 활성화된 에너지 물질의 경계면을 정확하게 추적하는 기법을 소해한다. 이러한 경계면 추적기법은 Level Set기반의 Ghost Fluid Concept을 이용하여 폭발과 같은 화학적 현상을 포함하는 에너지 물질의 경계면을 정확하게 파악하고, 쉽게 multi material, phase등과 같은 현상으로 확장이 가능하다. 본 연구에서는 이러한 multi physics modeling을 위한 선행연구로써 basic sharp interface tracking, impact problem에 대해서 소개할 것이다.

• International Journal of Control, Automation, and Systems
• /
• 제5권2호
• /
• pp.200-207
• /
• 2007

The passive direction finding cross localization method is widely adopted in passive tracking, therefore there will exist masses of false intersection points. Eliminating these false intersection points correctly and quickly is a key technique in passive localization. A new method is proposed for passive locating and tracking multi-jammer target in this paper. It not only solves the difficulty of determining the number of targets when masses of false intersection points existing, but also solves the initialization problem of elastic network. Thus this method solves the problem of multi-jammer target correlation and the elimination of static false intersection points. The method which dynamically establishes multiple hypothesis trajectory trees solves the problem of eliminating the remaining false intersection points. Simulation results show that computational burden of the method is lower, the elastic network can more quickly find all or most of the targets and have a more probability of locking the real targets. This method can eliminate more false intersection points.

• 정보보호학회논문지
• /
• 제21권6호
• /
• pp.141-152
• /
• 2011

다양한 감시 환경에서의 보안의 중요성이 대두됨에 따라 여러 대의 카메라로 움직이는 물체를 연속적으로 추적하는 시스템에 대한 연구가 활발히 진행되고 있다. 본 논문은 물체를 연속적으로 추적하기 위해 비겹침 다중 카메라 기반의 영삼감시시스템을 제안한다. 제안된 다중 IP 카메라 기반 객체추적 기술은 장치 간 hand-off 기술 및 프로토콜을 바탕으로 객체추적 모듈과 추적관리 모듈로 구성된다. 객체추적 모듈은 IP 카메라에서 실행되며 객체추적 정보 생성, 객체추적 정보 공유, 객체추적 정보를 이용한 객체 검색 및 모듈 내 설정 기능을 제공하고, 추적관리 모듈은 영상관제 서버에서 실행되며 객체추적 정보 실시간 수신, 객체추적 정보 검색, IP 카메라 컨트롤 기능을 제공한다. 본 논문에서 제안한 객체추적 기술은 다양한 감시 환경과 기술 방법에 의존하지 않는 범용적 프레임워크를 제안한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.947-948
• /
• 2008

Recently, many efforts have been made for research and application of object tracking system. However, introduced object tracking algorithms have limitations to adopt a realtime object tracking system with multi camera. In this paper, we present a novel background generation and target object recognition algorithm for realtime object tracking system with multi camera and implemented it.

• 한국컴퓨터정보학회논문지
• /
• 제22권7호
• /
• pp.47-54
• /
• 2017

Recently, correlation filter based trackers have shown excellent tracking performance and computational efficiency. In order to enhance tracking performance in the correlation filter based tracker, search area which is image patch for finding target must include target. In this paper, two methods to discriminatively represent target in the search area are proposed. Firstly, search area location is estimated using pyramidal Lucas-Kanade algorithm. By estimating search area location before filtering, fast motion target can be included in the search area. Secondly, we investigate multi-channel Local Edge Pattern(LEP) which is insensitive to illumination and noise variation. Qualitative and quantitative experiments are performed with eight dataset, which includes ground truth. In comparison with method without search area estimation, our approach retain tracking for the fast motion target. Additionally, the proposed multi-channel LEP improves discriminative performance compare to existing features.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.447-448
• /
• 2006

In this paper, we propose tracking algorithm using multi resolution in wavelet transform domain. This algorithm consists of two steps. The first step is feature extraction that is select feature-points using 1-level wavelet transform in ROI (Region of Interest). The other step is feature tracking. Based on multi resolution of wavelet transform, we estimate a displacement between current frame and next frame on the basis of selected feature-points. Experimental results show that the proposed algorithm confirmed a better performance than a centroid tracking and correlation tracking.

• 한국군사과학기술학회지
• /
• 제14권3호
• /
• pp.517-523
• /
• 2011

The estimation filter in radar systems must track targets' position within low tracking error. In the Multi-Function Radar(MFR), ${\alpha}-{\beta}$ filter and Kalman filter are widely used to track single or multiple targets. However, due to target maneuvering, these filters may not reduce tracking error, therefore, may lost target tracks. In this paper, a target tracking filter based on particle filtering algorithm is proposed for the MFR. The advantage of this method is that it can track targets within low tracking error while targets maneuver and reduce impoverishment of particles by the proposed resampling method. From the simulation results, the improved tracking performance is obtained by the proposed filtering algorithm.

• 전자공학회논문지S
• /
• 제36S권7호
• /
• pp.43-49
• /
• 1999

다중표적 추적시스템은 여러 개의 표적물을 동시에 추적한다. 이와 같은 시스템에서는 여러 개의 표적물들에 관한 위치정보들과 추적중인 표적물들과의 정보융합과정이 요구된다. 본 논문에서는 이러한 경우에 추적중인 표적물들이 지니는 예측위치들의 신뢰구간을 이용하여 측정한 위치정보들을 각각의 표적물들에 할당하는 방법을 제안하였다. 제안된 방법을 실제의 교통정보에 적용하여 그 우수한 특성을 살펴보았다.