• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.693-709
• /
• 2018

In this paper, we propose a compact Convolutional Neural Network (CNN)-based tracker in conjunction with a particle filter architecture, in which the CNN model operates as an accurate candidates estimator, while the particle filter predicts the target motion dynamics, lowering the overall number of calculations and refines the resulting target bounding box. Experiments were conducted on the Online Object Tracking Benchmark (OTB) [34] dataset and comparison analysis in respect to other state-of-art has been performed based on accuracy and precision, indicating that the proposed algorithm outperforms all state-of-the-art trackers included in the OTB dataset, specifically, TLD [16], MIL [1], SCM [36] and ASLA [15]. Also, a comprehensive speed performance analysis showed average frames per second (FPS) among the top-10 trackers from the OTB dataset [34].

• 한국지구물리탐사학회:학술대회논문집
• /
• 한국지구물리탐사학회 2005년도 제7회 특별심포지움 논문집
• /
• pp.11-18
• /
• 2005

Filament inversion routines are highly effective for target definition whenever total-field DHTEM vectors can be obtained using three-component logging tools. However most cross-hole components contain significant noise related to sensor design and errors in observation of probe rotation. Standard stacking methods can be used to improve data quality but additional statistical methods based on cross-correlation and spatial averaging of orthogonal components may be required to ensure a consistent vector migration path. Apart from assisting with spatial averaging, multiple filaments generated for successive time-windows can provide additional imaging information relating to target geometry and current migration. New digital receiver systems provide additional time-windows to provide better tracking options necessary for high-resolution imaging of this type.

• 대한임베디드공학회논문지
• /
• 제14권6호
• /
• pp.339-346
• /
• 2019

Multi-axis servo systems are widely used in various fields such as industiral systems for improving production efficiency, robotics and complex systems where many mechanical devices and sensor systems are connected. Such a servo system requires that the servo control technique to realize the synchronization of the drive shaft in the steady state and transient conditions and to control so as to follow the target track in order to improve product precision and production efficiency. In addition, embedded type hardware is required for smooth control of the entire multi-axis system. Therefore, this paper uses hardware based on FPGA which is widely used in digital signal processing field and various control system because hardware design change is easy and parallel processing is possible. In addition, Labview based servo motor control program was studied that can control the servo motor by ensuring the performance and flexibility of the FPGA and follow the target trajectory according to various speed processing and accurate timing synchronization.

• 한국공작기계학회논문집
• /
• 제14권4호
• /
• pp.13-21
• /
• 2005

The affected factors for an accuracy of theodolite system are not only the measurement environment of temperature, illumination, etc. but also the measurement processes of the distance between two theodolites, the distance from theodolite system to scale bar and the distance from theodolite system to targets. We have known that the best collimation distance between two theodolites and the best distance from theodolite system to scale bar is $3{\sim}4m$. This study was performed for searching the best distance from theodolite system to targets on above measurement configuration. And, we could know that the best distance from theodolite system to targets is $2{\sim}6m$ and the system accuracy could be within ${\pm}0.025mm$.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.1-7
• /
• 2019

Recently, smart factories have attracted much attention as a result of the 4th Industrial Revolution. Existing factory automation technologies are generally designed for simple repetition without using vision sensors. Even small object assemblies are still dependent on manual work. To satisfy the needs for replacing the existing system with new technology such as bin picking and visual servoing, precision and real-time application should be core. Therefore in our work we focused on the core elements by using deep learning algorithm to detect and classify the target object for real-time and analyzing the object features. We chose YOLO CNN which is capable of real-time working and combining the two tasks as mentioned above though there are lots of good deep learning algorithms such as Mask R-CNN and Fast R-CNN. Then through the line and inside features extracted from target object, we can obtain final outline and estimate object posture.

• 드라이브 ㆍ 컨트롤
• /
• 제17권4호
• /
• pp.8-14
• /
• 2020

Intelligent devices using ICT technology have been introduced in the field of construction machinery to improve productivity and stability. Among the intelligent devices, Machine Guidance is a device that provides real-time posture, location, and work range to drivers by installing various sensors, controllers, and satellite navigation systems on construction machines. Conversely, the efficiency of equipment that requires location information, such as machine guidance, will be greatly reduced in buildings, and tunnels in the GPS blind spots. Thus, the other high-precision positioning technologies are required in the GPS blind spot zone. In this study, we will develop a relative position measurement system that provides precise location information such as construction machinery and robots in a local area where the GPS reception is difficult. A relative position measurement system tracks a marker in the form of a sphere installed on a vehicle by using the image base tracking technology, and measures the distance and direction information to the marker to calculate a position.

• Journal of Astronomy and Space Sciences
• /
• 제22권1호
• /
• pp.13-20
• /
• 2005

지상 및 우주 천체 망원경용 비구면 반사경면 초기 제작공정에는 고정입자 휠을 사용하는 연삭이 있다. 본 연구에서는 매 연삭 가공 이전에 설정한 목표 표면조도를 달성할 수 있도록 입력 연삭변수들을 결정하고, 표면 가공오차를 추적하며 , 가공 경과시간을 최소화하는 새로운 연삭공정을 개발하였다. 특별히 이 공정 기법은 이전 연삭 가공 작업시 까지 수집된 입력 변수 및 가공 결과 표면조도 자료를 다 변수 회귀분석 방법에 대입하여 목표 표면조도에 따른 최적 연삭가공 입력변수를 매 가공 작업 시 진화적으로 제시하는 지능형 공정 조절 능력을 갖추고 있다. 개발된 공정기법과 초정밀 컴퓨터 수치제어 연삭기를 사용하여 $96.1\~65.0nm(Ra)$ 범위 의 목표 표면조도를 갖는 제로듀어 소재에 대하여 10회 가공 실험을 수행 한 결과 $=-0.906{\pm}3.38(\sigma)nm(Ra)$의 가공 정밀도를 달성하여, 지능형 연삭공정의 효율을 입증하였다. 이러한 연구결과는 천체망원경용 반사경면 연삭 가공 시 정성적 경험에 의존하여 가공하는 기존 기술을 극복하고 정량적 수치 모형에 의하여 가공소요시간 최소화 및 나노미터 급 표면조도를 달성하는 진화형 공정 최적화 기술의 확립이라는 의의를 가지고 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권1호
• /
• pp.272-287
• /
• 2016

Recently, various and effective contrast based salient object detection models to focus on a single target have been proposed. However, there is a lack of research on detection of multiple objects, and also it is a more challenging task than single target process. In the multiple target problem, we are confronted by new difficulties caused by distinct difference between properties of objects. The characteristic of existing models depending on the global maximum distribution of data point would become a drawback for detection of multiple objects. In this paper, by analyzing limitations of the existing methods, we have devised three main processes to detect multiple salient objects. In the first stage, regional features are extracted from over-segmented regions. In the second stage, the regional features are categorized into homogeneous cluster using the mean-shift algorithm with the kernel function having various sizes. In the final stage, we compute saliency scores of the categorized regions using only spatial features without the contrast features, and then all scores are integrated for the final salient regions. In the experimental results, the scheme achieved superior detection accuracy for the SED2 and MSRA-ASD benchmarks with both a higher precision and better recall than state-of-the-art approaches. Especially, given multiple objects having different properties, our model significantly outperforms all existing models.

• Journal of Positioning, Navigation, and Timing
• /
• 제6권2호
• /
• pp.53-57
• /
• 2017

eLoran refers to a terrestrial navigation system using high-power low-frequency signals. Thus, it can be regarded as a positioning, navigation and timing (PNT) system to back up a global navigation satellite system (GNSS) or an alternative to GNSS. South Korea is vulnerable to interference such as GNSS jamming in particular. Therefore, South Korea has made an effort to develop an independent navigation system through eLoran system. More particularly, an eLoran testbed has been developed to be used in the northwest sea area and research on applicability of eLoran in South Korea has been underway. The present study analyzes expected performance of eLoran according to locations of newly built eLoran transmitting stations as part of the eLoran testbed research. The performance of eLoran is analyzed in terms of horizontal position accuracy, and horizontal dilution of precision (HDOP) information was used since it affects accuracy significantly. The target service areas of the eLoran testbed are Incheon and Pyeongtaek Ports, and the required target performance is positioning accuracy of 20 m position within 30 km coverage of the target service area.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.940-943
• /
• 1996

In teleoperating, as seeing the monitor screen obtained from a camera instituted in the working environment, human operator generally controls the slave arm. Because we can see only 2-D image in a monitor, human operator does not know the depth information and can not work with high accuracy. In this paper, we proposed a traded control method using an visual sensor for the purpose of solving this problem. We can control a teleoperation system with precision when we use the proposed algorithm. Not only a human operator command but also an autonomous visual sensor feedback command is given to a slave arm for the purpose of coincidence current image features and target image features. When the slave arm place in a distant place from the target position, human operator can know very well the difference between the desired image features and the current image features, but calculated visual sensor command have big errors. And when the slave arm is near the target position, the state of affairs is changed conversely. With this visual sensor feedback, human does not need coincide the detail difference between the desired image features and the current image features and proposed method can work with higher accuracy than other method without, sensor feedback. The effectiveness of the proposed control method is verified through series of experiments.