• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• 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.

• Korean Journal of Family Social Work
• /
• no.23
• /
• pp.5-42
• /
• 2008

This research utilized the Grounded Theory and Giorgi's phenomenological research method to analyze the experience of battered women who are maintaining a non-violent marriage relationship. Due to the low accessibility of subjects and the distinctiveness of the topic, I have selected 4 participants and conducted in-depth interviews. As for data analysis results, there were 5 superior component elements and 19 inferior component elements for the experience of maintaining a non-violent marriage relationship and these appeared according to time. Thus, it can be analyzed as follows: 'the vicious cycle stage of violence: caught in a trap,' 'stage of breakaway from vicious violence cycle: making the self-rescue measures,' 'entry to a new track: retrieval of autonomy,' 'stage of forming a non-violence track: preventing the recurrence of the violence,' 'stage of maintaining non-violent track: conversion to non-violent relationship.' This research result will be useful in seeking an effective social welfare intervention plan for successful non-violent relationship in order to help 50% of the battered women from family violence who wish to maintain marriage.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.667-674
• /
• 2022

In this study, the correlation between the damage type and operating conditions of the sleepers was analyzed based on the design data and visual inspection results for the concrete sleepers of the sleeper floating track (STEDEF) that have been in operation for more than 20 years. It appeared in the form of cracks, breakages, and breaks in the concrete at the center and tie bar contact and buried areas. As a result of the numerical analysis, it was analyzed that the change in the left and right spring stiffness of the sleeper resilience pad increases the maximum stress, tensile stress, compressive stress, and displacement of the concrete sleeper, and stress concentration in the concrete at the tie bar contact area. It was proved analytically that the sleeper resilience pad can affect the damage of the concrete sleeper. Therefore, damage of concrete sleepers in the sleeper floating track in urban transit could be caused by changes in spring stiffness of sleeper resilience pads. It was reviewed that preventive maintenance such as improvement and timely replacement of sleeper resilience pads was necessary.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.11
• /
• pp.1443-1450
• /
• 2019

The naval combat system is gradually evolving, actively accepting technological advances. The domain components of the naval combat system must also be adapted to the gradual evolution of the naval battle system. In other words, domain components should be easy to change. However, existing domain components do not fully consider changeability. In particular, when receiving more messages than planned from the middleware, it does not consider a way to process them in a timely manner. We propose a domain component reference architecture for timely processing of increased messages. The proposed architecture separates domain components so that business logic can operate independently, and sets up multiple business logic to operate simultaneously to handle increased messages. The track management domain component is designed using the proposed reference architecture to confirm the applicability.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.10
• /
• pp.225-231
• /
• 2009

This paper describes a new algorithm to compute and track the deformation of an isosurface component defined in a time-varying volumetric data. Isosurface visualization is one of the most common method for effective visualization of volumetric data. However, most isosurface visualization algorithms have been developed for static volumetric data. As imaging and simulation techniques are developed, large time-varying volumetric data are increasingly generated. Hence, development of time-varying isosurface visualization that utilizes dynamic properties of time-varying data becomes necessary. First, we define temporal correspondence between isosurface components of two consecutive timesteps. Based on the definition, we perform an algorithm that tracks the deformation of an isosurface component that can be selected using the Contour Tree. By repeating this process for entire timesteps, we can effectively visualize the time-varying data by displaying the dynamic deformation of the selected isosurface component.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.11
• /
• pp.11-21
• /
• 1996

Robot manipulators constituing multi-robot system must exert the desired motion force on an object to preserve tghe fine motion of it. The forces exerte on an object by the end-effectors of multi-inducing force and the internal force. Here, motion-inducing force effects the motion of an object, but internal force as lies in the null space of an object coordinate can't effect it. The motion of an object can't track exactly the desired motion by the effect of an object, but internal force as lies in the null space of the effect of internal force component, therefore internal force component must be considered. In this paper, first, under assumption that we can estimate exactly the parameter of dynamics, we constitute paper, first, under assumption that we can estimate exactly the parameter of dynamics, we constitute the controller concerning internal force. And we obtain the internal force as projecting force sensor readings onto the space spanned by null basis set of jacobian matrix. Using the resolved acceleration control method and the fact that internal force lies in the null space of jacobian matrix, we construct the robust control law to preserve the robustness with respect to the uncertainty of mainpulator parameters.

• Journal of Power Electronics
• /
• v.14 no.6
• /
• pp.1100-1108
• /
• 2014

Modular multilevel converters (MMCs) have emerged as the most promising topology for high and medium voltage applications for the coming years. However, one particular negative characteristic of MMCs is the existence of circulating current, which contains a dc component and a series of low-frequency even-order ac harmonics. If not suppressed, these ac harmonics will distort the arm currents, increase the power loses, and cause higher current stresses on the semiconductor devices. Repetitive control (RC) is well known due to its distinctive capabilities in tracking periodic signals and eliminating periodic errors. In this paper, a novel circulating current control scheme base on RC is proposed to effectively track the dc component and to restrain the low-frequency ac harmonics. The integrating function is inherently embedded in the RC controller. Therefore, the proposed circulating current control only parallels the RC controller with a proportional controller. Thus, conflicts between the RC controller and the traditional proportional integral (PI) controller can be avoided. The design methodologies of the RC controller and a stability analysis are also introduced. The validity of the proposed circulating current control approach has been verified by simulation and experimental results based on a three-phase MMC downscaled prototype.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.5
• /
• pp.1240-1250
• /
• 2010

In this paper, an eye tracking method is presented by using on iterated spatial moment adapting weighted gray level that can accurately detect and track user's eyes under the complicated background. The region of face is detected by using Haar-like feature before extracting region of eyes to minimize an region of interest from the input picture of CCD camera. And the region of eyes is detected by using eigeneye based on the eigenface of Principal component analysis. Also, feature points of eyes are detected from darkest part in the region of eyes. The tracking of eyes is achieved correctly by using iterated spatial moment adapting weighted gray level.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.198-201
• /
• 2009

In this paper, an eye tracking method is presented by using on iterated spatial moment adapting weighted gray level that can accurately detect and track user's eyes under the complicated background. The region of face is detected by using Haar-like feature before extracting region of eyes to minimize an region of interest from the input picture of CCD camera. And the region of eyes is detected by using eigeneye based on the eigenface of Principal component analysis. And then feature points of eyes are detected from darkest part in the region of eyes. The tracking of eyes is achieved correctly by using iterated spatial moment adapting weighted gray level.

• Smart Structures and Systems
• /
• v.24 no.3
• /
• pp.319-332
• /
• 2019

Structural health monitoring (SHM) is of great importance to super high-rise buildings. The Shanghai Tower is currently the tallest building in China, and a complete SHM system was simultaneously constructed at the beginning of the construction of the tower. Due to the variety of sensor types and the large number of measurement points in the SHM system, an online automatic structural health assessment method with few computations and no manual intervention is needed. This paper introduces a structural health assessment method for the Shanghai Tower that uses the coefficients of an autoregressive (AR) time series model as structural state indicators. An analysis of collected data indicates that the coefficients of the AR model are affected by environmental factors, and the principal component analysis method is used to remove the influence of environmental factors. Finally, the control chart method is used to track the changes in structural state indicators, and a plan for online automatic structure health state evaluation is proposed. This method is applied to long-term acceleration and inclination data from the Shanghai Tower and successfully identifies the changes in the structural state. Overall, the structural state indicators of the Shanghai Tower are stable, and the structure is in a healthy state.