• Journal of Korea Water Resources Association
• /
• v.47 no.4
• /
• pp.357-370
• /
• 2014

This study aims to develop a rainfall field tracking method for depth-area-duration (DAD) analysis and assess whether the proposed tracking methods are able to properly estimate the maximum average areal rainfall (MAAR) within the study area during a rainfall period. We proposed three different rainfall field tracking algorithms (Box-tracking, Point-tracking, Advanced point-tracking) and then applied them to the virtual rainfall field with 1hr duration and also compared DAD curves of each method. In addition, we applied the three tracking methods and a traditional GIS-based tool to the typhoon 'Nari' rainfall event of the Yongdam-Dam watershed and then assess applicability of the proposed methods for DAD analysis. The results showed that Box-tracking was much faster than the other two tracking methods in terms of searching for the MAAR but it was impossible to describe rainfall spatial pattern during its tracking processes. On the other hand, both Point-tracking and Advanced point-tracking provided the MAAR by considering the spatial distribution of rainfall fields. In particular, Advanced point-tracking estimated the MAAR more accurately than Point-tracking in the virtual rainfall field, which has two rainfall centers with similar depths. The proposed automatic rainfall field tracking methods can be used as effective tools to analyze DAD relationship and also calculate areal reduction factor.

• Journal of Advanced Marine Engineering and Technology
• /
• v.30 no.3
• /
• pp.438-443
• /
• 2006

The probe type sensor using strain gauges was used to track a container box's seam in this paper. A strain gauge has a property which the specific resistance is changed by varying of the sectional area and length when tensile and compressive stresses are generated at the strain gauge. We designed the automatic seam tracking device by attaching the probe type strain gauge sensor, motor driving slide, encorder to check the moving distance and interface card connected MPU upside the speed controllable carriage. The folded work piece for a container box is made to examine whether the device can track the seam automatically or not. Seam tracking experiments were done by changing the carriage moving speed at 300. 400. 500. 600[mm/min] each as the voltage of side track was 2.5[V]. We compared and analyzed the sampling data which is obtained by output voltage of strain gauge sensor and rotary encorder pulse every 100 [m/s]. The welding experiments were performed by using $CO_2$ welding machine about the carriage moving speed that has good seam tracking condition in the seam tracking experiments above. And we compared the seam tracking status.

• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.305-311
• /
• 2021

One of the highly reliable object tracking methods is to trace objects by associating objects detected by deep learning. The detected object is represented by a rectangular box. The box has information such as location and size. Since the tracker has motion information of the object in addition to the location and size, knowing additional information about the motion of the detected box can increase the reliability of object tracking. In this paper, we present a new method of reliably estimating the moving direction of the detected object in underground parking lot. First, the frame difference image is binarized for detecting motion energy, change due to the object motion. Then, a cumulative binary image is generated that shows how the motion energy changes over time. Next, the moving direction of the detected box is estimated from the accumulated image. We use a new cost function to accurately estimate the direction of movement of the detected box. The proposed method proves its performance through comparative experiments of the existing methods.

• Journal of Satellite, Information and Communications
• /
• v.8 no.1
• /
• pp.77-82
• /
• 2013

The object tracking method using particle filtering has been proved successful since it is based on the Monte Carlo simulation to estimate the posterior distribution of the state vector that is nonlinear and non-Gaussian in the real-world situation. In this paper, we present two nobel methods that can improve the performance of the object tracking algorithm based on the particle filtering. First one is the feedback method that replace the low-weighted tracking sample by the estimated state vector in the previous frame. The second one is an tracking box correction method to find an confidence interval of back projection probability on the estimated candidate object area. An sample propagation equation is also presented, which is obtained by experiments. We designed well-organized test data set which reflects various challenging circumstances, and, by using it, experimental results proved that the proposed methods improves the traditional particle filter based object tracking method.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.18 no.2
• /
• pp.16-22
• /
• 2010

This paper proposes the real-time moving object tracking system UAV using color information. Case of object tracking, it have studied to recognizing the moving object or moving multiple objects on the fixed camera. And it has recognized the object in the complex background environment. But, this paper implements the moving object tracking system using the pan/tilt function of the camera after the object's region extraction. To do this tracking system, firstly, it detects the moving object of RGB/HSI color model and obtains the object coordination in acquired image using the compact boundary box. Secondly, the camera origin coordination aligns to object's top&left coordination in compact boundary box. And it tracks the moving object using the pan/tilt function of camera. It is implemented by the Labview 8.6 and NI Vision Builder AI of National Instrument co. It shows the good performance of camera trace in laboratory environment.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.5
• /
• pp.281-288
• /
• 2022

According to statistics conducted by the Korea Coast Guard, the number of accidents on small boats under 5 tons is increasing every year. This is because only a small number of people are on board. The previously developed maritime distress and safety systems are not well distributed because passengers must be equipped with additional remote equipment. The purpose of this study is to develop a distress awareness system that recognizes man over-board situations in real time. This study aims to present the part of the passenger tracking system among the small ship's distress awareness situational system that can generate passenger's location information in real time using deep learning based object detection and tracking technologies. The system consisted of the following steps. 1) the passenger location information is generated in the form of Bounding box using its detection model (YOLOv3). 2) Based on the Bounding box data, Deep SORT predicts the Bounding box's position in the next frame of the image with Kalman filter. 3) When the actual Bounding Box is created within the range predicted by Kalman-filter, Deep SORT repeats the process of recognizing it as the same object. 4) If the Bounding box deviates the ship's area or an error occurs in the number of tracking occupant, the system is decided the distress situation and issues an alert. This study is expected to complement the problems of existing technologies and ensure the safety of individuals aboard small boats.

• Journal of Electrical Engineering and Technology
• /
• v.6 no.3
• /
• pp.391-396
• /
• 2011

The tracking pattern formed on the dielectric surface due to a surface electrical discharge exhibits fractal structure. In order to quantitatively investigate the fractal characteristics of the surface tracking pattern, the dielectric breakdown model has been employed to numerically generate the surface tracking pattern. In dielectric breakdown model, the pattern growth is determined stochastically by a probability function depending on the local electric potential difference. For the computation of the electric potential for all points of the lattice, a two-dimensional discrete Laplace equation is solved by mean of the successive over-relaxation method combined to the Gauss-Seidel method. The box counting method has been used to calculate the fractal dimensions of the simulated patterns with various exponent $\eta$ and breakdown voltage $\phi_b$. As a result of the simulation, it is found that the fractal nature of the surface tracking pattern depends strongly on $\eta$ and $\phi_b$.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.18-20
• /
• 2015

본 논문은 마그네틱 센서 기반의 오브젝트 추적 기술을 이용한 혼합현실 선거 방송 시스템을 제안한다. 마커 기반의 증강현실 기술은 방송환경에서는 강한 조명으로 인하여 마커의 특징점 추출의 간섭 및 소실로 추적이 끊기는 문제가 있다. 특히, 선거방송와 같은 생방송 중에 그래픽이 튀거나 사라지는 것은 방송 사고와 다름없다. 따라서, 우리는 조명이나 가림의 영향 없이 추적 성능을 강인하게 하기 위해서, 무선의 마그네틱 센서를 내장한 별도로 제작한 투표함 모형을 추적하도록 하였다. 본 논문에서는 마그네틱 센서를 내장한 실물 투표함을 실시간으로 추적하게 하고, 그 정보를 증강현실 방송 시스템과 통합한 시스템 구성 및 그것을 이용한 증강현실 선거 방송 어플리케이션을 보여준다. 그 결과, 연기자가 선거정보그래픽과 연동하는 투표함을 자유롭고 직관적으로 움직일 수 있었으며, 자연스러운 증강현실 합성 결과를 얻을 수 있었다.

• Journal of Advanced Navigation Technology
• /
• v.23 no.1
• /
• pp.77-83
• /
• 2019

In this paper, we propose an object tracking algorithm based on color histogram and convolutional neural network model. In order to increase the tracking accuracy, we synthesize generic object tracking using regression network algorithm which is one of the convolutional neural network model-based tracking algorithms and a mean-shift tracking algorithm which is a color histogram-based algorithm. Both algorithms are classified through support vector machine and designed to select an algorithm with higher tracking accuracy. The mean-shift tracking algorithm tends to move the bounding box to a large range when the object tracking fails, thus we improve the accuracy by limiting the movement distance of the bounding box. Also, we improve the performance by initializing the tracking start positions of the two algorithms based on the average brightness and the histogram similarity. As a result, the overall accuracy of the proposed algorithm is 1.6% better than the existing generic object tracking using regression network algorithm.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.1
• /
• pp.9-15
• /
• 2023

As mobile devices such as smartphones, tablets, and kiosks become increasingly prevalent, there is growing interest in developing alternative input systems in addition to traditional tools such as keyboards and mouses. Many people use their own bodies as a pointer to enter simple information on a mobile device. However, methods using the body have limitations due to psychological factors that make the contact method unstable, especially during a pandemic, and the risk of shoulder surfing attacks. To overcome these limitations, we propose a simple information input system that utilizes gaze-tracking technology to input passwords and control web surfing using only non-contact gaze. Our proposed system is designed to recognize information input when the user stares at a specific location on the screen in real-time, using intelligent gaze-tracking technology. We present an analysis of the relationship between the gaze input box, gaze time, and average input time, and report experimental results on the effects of varying the size of the gaze input box and gaze time required to achieve 100% accuracy in inputting information. Through this paper, we demonstrate the effectiveness of our system in mitigating the challenges of contact-based input methods, and providing a non-contact alternative that is both secure and convenient.