• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.11
• /
• pp.789-794
• /
• 2013

A vehicle tracking system makes it possible to induce the vehicle movement path for avoiding traffic congestion and to prevent traffic accidents in advance by recognizing traffic flow, monitoring vehicles, and detecting road accidents. To track the vehicles effectively, those which appear in a sequence of video frames need to identified by extracting the features of each object in the frames. Next, the identical vehicles over the continuous frames need to be recognized through the matching among the objects' feature values. In this paper, we identify objects by binarizing the difference image between a target and a referential image, and the labelling technique. As feature values, we use the center coordinate of the minimum bounding rectangle(MBR) of the identified object and the averages of 1D FFT(fast Fourier transform) coefficients with respect to the horizontal and vertical direction of the MBR. A vehicle is tracked in such a way that the pair of objects that have the highest similarity among objects in two continuous images are regarded as an identical object. The experimental result shows that the proposed method outperforms the existing methods that use geometrical features in tracking accuracy.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2254-2259
• /
• 2005

In this paper, we propose a wavelet based fuzzy neural network(WFNN) based direct adaptive control scheme for the solution of the tracking problem of mobile robots. To design a controller, we present a WFNN structure that merges advantages of neural network, fuzzy model and wavelet transform. The basic idea of our WFNN structure is to realize the process of fuzzy reasoning of wavelet fuzzy system by the structure of a neural network and to make the parameters of fuzzy reasoning be expressed by the connection weights of a neural network. In our control system, the control signals are directly obtained to minimize the difference between the reference track and the pose of mobile robot using the gradient descent(GD) method. In addition, an approach that uses adaptive learning rates for the training of WFNN controller is driven via a Lyapunov stability analysis to guarantee the fast convergence, that is, learning rates are adaptively determined to rapidly minimize the state errors of a mobile robot. Finally, to evaluate the performance of the proposed direct adaptive control system using the WFNN controller, we compare the control performance of the WFNN controller with those of the FNN, the WNN and the WFM controllers.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.9
• /
• pp.147-152
• /
• 2015

Almost all existing inspection methods for catenary system have relied on dedicated track inspection cars. Two factors make it necessary to modify the conventional method: First, the climate has become increasingly similar to that of a subtropical area. In addition, high speed trains have been developed that can run as fast as 450Km/h. This paper presents a visual catenary inspection system structure that incorporates roadbed information for high-speed running environments. Conventional catenary inspection systems determined the degree of wearing only for the trolley wire. The facility inspection system presented in this paper on the other hand utilizes visual and sound data of inspection targets and adjusts inspection frequency and sensitivity depending on the geographic data.

• Journal of Korea Multimedia Society
• /
• v.16 no.5
• /
• pp.619-628
• /
• 2013

In this paper, we describe a strongly improved tracking method using combination of CAMshift and Kalman filter algorithm. CAMshift algorithm doesn't consider the object's moving direction and velocity information when it set the search windows for tracking. However if Kalman filter is combined with CAMshift for setting the search window, it can accurately predict the object's location with the object's present location and velocity information. By using this prediction before CAMshift algorithm, we can track fast moving objects successfully. Also in this research, we show better tracking results than conventional approaches which make use of single color information by using both color information of HSV and YCrCb simultaneously. This modified approach obtains more robust color segmentation than others using single color information.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.30-34
• /
• 1998

Path Planning is one of the important fields in robot technologies. Local path planning may be done in on-line modes while recognizing an environment of robot by itself. In dynamic environments to obtain fluent information for environments vision system as a sensing equipment is a one of the most necessary devices for safe and effective guidance of robots. If there is a predictor that tells what future sensing outputs will be, robot can respond to anticipated environmental changes in advance. The tracking of obstacles has a deep relationship to the prediction for safe navigation. We tried to deal with active contours, that is snakes, to find out the possibilities of stable tracking of objects in image plane. Snakes are defined based on energy functions, and can be deformed to a certain contour form which would converge to the minimum energy states by the forces produced from energy differences. By using point algorithm we could have more speedy convergence time because the Brent's method gives the solution to find the local minima fast. The snake algorithm may be applied to sequential image frames to track objects in the images by these characteristics of speedy convergence and robust edge detection ability.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.5
• /
• pp.505-511
• /
• 2011

In this paper, we have developed an automatic velocity control system of a small-sized commercial treadmill (belt length of 1.2 m and width of 0.5 m) which is widely used at home and health centers. The control objective is to automatically adjust the treadmill velocity so that the subject's position is maintained within the track when the subject walks at a variable velocity. The subject's position with respect to a reference point is measured by a low-cost sonar sensor located on the back of the subject. Based on an encoder sensor measurement at the treadmill motor, a state feedback control algorithm with Kalman filter was implemented to determine the velocity of the treadmill. In order to reduce the unnatural inertia force felt by the subject, a predefined acceleration limit was applied, which generated smooth velocity trajectories. The experimental results demonstrate the effectiveness of the proposed method in providing successful velocity changes in response to variable velocity walking without causing significant inertia force to the subject. In the pilot study with three subjects, users could change their walking velocity easily and naturally with small deviations during slow, medium, and fast walking. The proposed automatic velocity control algorithm can potentially be applied to any locomotion interface in an economical way without having to use sophisticated and expensive sensors and larger treadmills.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.20 no.2
• /
• pp.165-170
• /
• 2002

For the inaccessible area where the field verification is unable, it is difficult to obtain the ground control points (GCPs) or the acquired GCPs may be inaccurate. In general systematic geometric correction is achieved by utilizing orbit ephemeris and three axis attitude data of the satellite. however, this method results to poor accuracy of the imagery's absolute coordinates. To improve the absolute accuracy as well as the relative accuracy, we added the accessible region into the inaccessible area. We obtained GCPs in the accessible region by the fast static GPS survey and made geometric corrections with these biased GCPs. Because the biased GCPs show a pattern of coordinate errors, we analyzed this tendency to track the estimated errors in the inaccessible area.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.4
• /
• pp.1100-1112
• /
• 1998

In this paper, a novel method of minimizing the phase error is proposed. A DP-PLL system using this method is implemented and its performacnce is investigated, too. The DP-PLL system detects the phase error between reference clock and locally generated system clock. The phase difference is then reported as a PEV(Phase Error Variation), which is propoced from the delayted n-tap rising dege clock circuit with 5ns resolution in the phase detector. The algorithm is used to track the optimal DAC coefficients, which are adjusted from sample to sample in such a way as to minimize the PEV. The proposed method is found to have remarkable good potential for fast and accurate phase error tracking characteristic. The algorithm shows good performance to supress the low frequency jitter.-ending points, we design new basis functions based on the Legendre polynomial and then transform the error signals with them. When applied to synthetic images such as circles, ellipses and etc., the proposed method provides, in overall, outstanding results in respect to the transform coding gain compared with DCT and DST. And in the case when applied to natural images, the proposed method gives better image quality over DCT and comparable results with DST.

• The Journal of the Korea Contents Association
• /
• v.17 no.3
• /
• pp.42-52
• /
• 2017

Panorama image is a single image obtained by combining images taken at several viewpoints through matching of corresponding points. Existing panoramic image generation methods that find the corresponding points are extracting local invariant feature points in each image to create descriptors and using descriptor matching algorithm. In the case of video sequence, frames may be a lot, so therefore it may costs significant amount of time to generate a panoramic image by the existing method and it may has done unnecessary calculations. In this paper, we propose a method to quickly create a single panoramic image from a video sequence. By assuming that there is no significant changes between frames of the video such as in locally, we use the FAST algorithm that has good repeatability and high-speed calculation to extract feature points and the Lucas-Kanade algorithm as each feature point to track for find the corresponding points in surrounding neighborhood instead of existing descriptor matching algorithms. When homographies are calculated for all images, homography is changed around the center image of video sequence to warp images and obtain a planar panoramic image. Finally, the spherical panoramic image is obtained by performing inverse transformation of the spherical coordinate system. The proposed method was confirmed through the experiments generating panorama image efficiently and more faster than the existing methods.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.3
• /
• pp.501-511
• /
• 2022

In 2021, ransomware attacks became popular, and the number is rapidly increasing every year. Since PowerShell is used as the primary ransomware technique, the need for PowerShell-based malware detection is ever increasing. However, the existing detection techniques have limits in that they cannot detect obfuscated scripts or require a long processing time for deobfuscation. This paper proposes a simple and fast deobfuscation method and a deep learning-based classification model that can detect PowerShell-based malware. Our technique is composed of Word2Vec and a convolutional neural network to learn the meaning of a script extracting important features. We tested the proposed model using 1400 malicious codes and 8600 normal scripts provided by the AI-based PowerShell malicious script detection track of the 2021 Cybersecurity AI/Big Data Utilization Contest. Our method achieved 5.04 times faster deobfuscation than the existing methods with a perfect success rate and high detection performance with FPR of 0.01 and TPR of 0.965.