• ETRI Journal
• /
• v.42 no.3
• /
• pp.411-419
• /
• 2020

In real-world intelligent transportation systems, accuracy in vehicle license plate detection and recognition is considered quite critical. Many algorithms have been proposed for still images, but their accuracy on actual videos is not satisfactory. This stems from several problematic conditions in videos, such as vehicle motion blur, variety in viewpoints, outliers, and the lack of publicly available video datasets. In this study, we focus on these challenges and propose a license plate detection and recognition scheme for videos based on a temporal matching prior network. Specifically, to improve the robustness of detection and recognition accuracy in the presence of motion blur and outliers, forward and bidirectional matching priors between consecutive frames are properly combined with layer structures specifically designed for plate detection. We also built our own video dataset for the deep training of the proposed network. During network training, we perform data augmentation based on image rotation to increase robustness regarding the various viewpoints in videos.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.467-467
• /
• 2000

In this paper, a structural design method of robust motion controllers for high-accuracy positioning systems, which makes it possible to predict the performance of the whole closed-loop system, is proposed. First, a stabilizing control input is designed based on robust internal-loop compensate.(RTC) for the system in the presence of uncertainty and disturbance. Next, using the structural characteristics of the RIC, disturbance attenuation properties and the performance of the closed-loop system determined by the variation of controller gains are analyzed. Through this analysis, in some specific applications, it is shown that if the control gain of RIC is increased by N times, the magnitude of error is reduced to its 1/N. Finally, the proposed method is verified through experiments using a high-accuracy positioning system used in the semiconductor chip mounting devices.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.11a
• /
• pp.449-450
• /
• 2010

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.672-676
• /
• 2004

In this paper, measuring system of 5 DOF motion errors are proposed using two capacitive type sensor, a straight edge and a laser interfoerometer. Yawing error and pitching error are measured using the laser interferometer, and rolling error is measured by the reversal method using a capacitive type sensor. Linear motion errors of horizontal and vertical direction are measured using the sequential two point method. In this case, influence of angular motion errors is compensated using the previously measured angular motion errors. In the horizontal direction, measuring accuracy is within 0.05 $\mu$m and 0.27 arcsec, and in the vertical direction, it is within 0.15 $\mu$m and 0.5 arcsec. From these results, it is confirmed that the proposed measureing system is very effective to the measurement of 5 DOF motion errors in the ultra precision feed tables.

• Journal of Sensor Science and Technology
• /
• v.5 no.4
• /
• pp.57-70
• /
• 1996

Motion can be estimated continuously from each sensor through the analysis of the instantaneous states of an object. This paper is aimed to introduce a method to estimate the general 3D motion of a planar object from the instantaneous states of an object using multi-sensor data fusion. The instantaneous states of an object is estimated using the linear feedback estimation algorithm. The motion estimated from each sensor is fused to provide more accurate and reliable information about the motion of an unknown planar object. We present a fusion algorithm which combines averaging and deciding. With the assumption that the motion is smooth, the approach can handle the data sequences from multiple sensors with different sampling times. Simulation results show proposed algorithm is advantageous in terms of accuracy, speed, and versatility.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.2
• /
• pp.275-280
• /
• 2004

In this paper, a very fast block matching scheme was proposed to reduce the computations of Block Sum Pyramid Algorithm for motion estimation in video coding. The proposed algorithm is based on Block Sum Pyramid Algorithm and Efficient Multi-level Successive Elimination Algorithm. The proposed algorithm can reduce the computations of motion estimation greatly with 100% motion estimation accuracy. The efficiency of the proposed algorithm was verified by experimental results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.2C
• /
• pp.141-146
• /
• 2003

In this paper, we propose an adaptive search range decision algorithm for motion vector estimation in video coding. The performance of general motion estimation method in video coding mechanism is evaluated with respect to the motion vector accuracy and the complexity, which is trade-off. The proposed algorithm that plays as a role of pre-processing for motion vector estimation determines the motion search range by the local statistics of motion vector of neighboring blocks, resulting in more than 60(%) reduction of the computational cost without the loss of visual quality. Experimental results show the capability of the proposed algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.6
• /
• pp.131-140
• /
• 2017

As video equipments such as CCTV are used for various purposes in fields of society, digital video data processing technology such as automatic motion detection is essential. In this paper, we proposed and implemented a more stable and accurate motion detection system based on background subtraction technique. We could improve the accuracy and stability of motion detection over existing methods by efficiently processing color information of digital image data. We divided the procedure of color information processing into each components of color information : brightness component, color component of color information and merge them. We can process each component's characteristics with maximum consideration. Our color information processing provides more efficient color information in motion detection than the existing methods. We improved the success rate of motion detection by our background update process that analyzed the characteristics of the moving background in the natural environment and reflected it to the background image.

• Progress in Medical Physics
• /
• v.28 no.1
• /
• pp.1-10
• /
• 2017

The difference between three-dimensional (3D) and four-dimensional (4D) dose could be affected by factors such as tumor size and motion. To quantitatively analyze the effects of these factors, a phantom that can independently control each factor is required. The purpose of this study is to develop a deformable lung phantom with the above attributes and evaluate the characteristics. A phantom was designed to simulate diaphragm motion with amplitude in the range 1~7 cm and period up to ${\geq}2s$ of regular breathing. To simulate different tumors sizes, custom molds were created using a 3D printer and filled with liquid silicone. The accuracy of the phantom diaphragm motion was assessed by comparing measured motion with predicted motion. Because the phantom diaphragm motion is not identical to the tumor motion, the correlation between the diaphragm and tumor motions was calculated by a curve fitting method to emulate user-intended tumor motion. Tumors of different sizes were located at same position, and tumor set-up positions were evaluated. The accuracy of phantom diaphragm motion was better than 1 mm. The diaphragm-tumor correlation showed that the tumor motion in the superior-inferior direction increased with increasing diaphragm motion. The tumor motion was larger in the $10cm^3$ tumor than in the $90cm^3$ tumor. The range of difference between the tumor set-up positions was 0 to 0.45 cm. This phantom showed independently adjusting factors such as tumor size and motion to facilitate quantitative analysis of the dosimetric impact of respiratory motion according to these factors.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1086-1092
• /
• 2009

Stereoscopic conversion has been an important and challenging issue for many 3-D video applications. Usually, there are two different stereoscopic conversion approaches, i.e., image motion-based conversion that uses motion information and object-based conversion that partitions an image into moving or static foreground object(s) and background and then converts the foreground in a stereoscopic object. As well, since the input sequence is MPEG-1/2 compressed video, motion data stored in compressed bitstream are often unreliable and thus the image motion-based conversion might fail. To solve this problem, we present the utilization of key-motion that has the better accuracy of estimated or extracted motion information. To deal with diverse motion types, a transform space produced from motion vectors and color differences is introduced. A key-motion is determined from the transform space and its associated stereoscopic image is generated. Experimental results validate effectiveness and robustness of the proposed method.