• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.425-430
• /
• 2015

This paper is a study to improve the classification efficiency of rotating objects by using deep neural networks to which a deep learning algorithm was applied. For the classification experiment of rotating objects, COIL-20 is used as data and total 3 types of classifiers are compared and analyzed. 3 types of classifiers used in the study include PCA classifier to derive a feature value while reducing the dimension of data by using Principal Component Analysis and classify by using euclidean distance, MLP classifier of the way of reducing the error energy by using error back-propagation algorithm and finally, deep learning applied DBN classifier of the way of increasing the probability of observing learning data through pre-training and reducing the error energy through fine-tuning. In order to identify the structure-specific error rate of the deep neural networks, the experiment is carried out while changing the number of hidden layers and number of hidden neurons. The classifier using DBN showed the lowest error rate. Its structure of deep neural networks with 2 hidden layers showed a high recognition rate by moving parameters to a location helpful for recognition.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.12 no.6
• /
• pp.1159-1166
• /
• 2017

Cattle behavior detection system based on Internet of Things is designed and implemented by using gyroscope and acceleration sensor, Arduino pro-mini and bluetooth module. The implemented system measures cattle's moving and the measured data are transmitted to smart phone by bluetooth module. They are displayed by 2-dimensional graph on the smart phone and the number of cattle's step are calculated from the graph. The detected and gathered data from the system are analyzed by the proposed algorithm to decide which cows are in the estrus or not, and the proposed system can be used to increase the success rate of artificial insemination in normal estrus by detecting cow's behaviors such as the number of steps and jumping. In this paper, the implemented cattle behavior detecting system are strapped on cattle's leg and it measures cattle behaviors for determining that a cattle is estrus or not by the proposed algorithm. In the future research, the system which lengthens communication distance and increases the number of cattle under the test will be considered and also the measured data will be database for cattle research.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.717-719
• /
• 2015

The research and development of the unmanned vehicle is being carried out actively in domestic and foreign countries. The research is being carried out to provide various services so that the weakness of system such as conventional 2D-based navigation systems can be supplemented and the driving can be safer. This paper suggests the method that enables real-time video processing in more efficient way by realizing the location detection and signal recognition technique of traffic signals in video. In order to overcome the limit of conventional methods that have a difficulty in analyzing the signal as it is sensitive to brightness change, the proposed method realizes the program that grasps the depth data in front of the vehicle using video processing, analyzes the signal by detecting traffic signal and estimates color components of traffic signal in front and the distance between traffic signal and the vehicle.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.326-327
• /
• 2016

A study on Ontology-based indoor positioning techniques using BLE Beacon. Recently BLE beacon has been widely used as a technique for measuring the indoor location. But it requires a filtering technique for the measurement of the correct position, and uses the most fixed beacon. It is not accurate that calculates the position information through the identification of the beacon signal. Therefore, filtering is important. So it takes a lot of time, position measurement and filtering. Thus, we is to measure the exact position at the indoor using a mobile beacon. The measured beacon signal is composed of an ontology for reuse in the same pattern. RSSI is measured the receiver is the distance of the beacon. And this value configure the location ontology to be normalized by the relationship analysis between the values. The ontology is a method for calculating the position information of the moving beacon. It may be detected fast and accurate indoor position information.

• Journal of Digital Contents Society
• /
• v.18 no.5
• /
• pp.989-994
• /
• 2017

In this paper, we propose a pothole detection system using 2D LiDAR and a pothole detection algorithm. Conventional pothole detection methods can be divided into vibration-based method, 3D reconstruction method, and vision-based method. Proposed pothole detection system uses two inexpensive 2D LiDARs and improves pothole detection performance. Pothole detection algorithm is divided into preprocessing for noise reduction, clustering and line extraction for visualization, and gradient function for pothole decision. By using gradient of distance data function, we check the existence of a pothole and measure the depth and width of the pothole. The pothole detection system is developed using two LiDARs, and the 3D pothole detection performance is shown by detecting a pothole with moving LiDAR system.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.9
• /
• pp.877-880
• /
• 2009

A curing process in post-printing section of R2R process is required for an electrical property of the printed pattern when devices such as RFID, Solar cell are printed. PEN as well as heat-stabilized PET which is used as a plastic substrate would be deformed at high temperature due to change of its elastic modulus. And crack in the printed pattern, which is on the plastic substrate is occurred due to the deformation of the substrate. The occurrence of crack causes electrical resistance to increase and the quality of the device to deteriorate. In case of RFID antenna, the range of reading distance is shortened as the electrical resistance of the antenna is increased. Therefore, the deformation of the plastic substrate, which causes the occurrence of crack, should be minimized by setting up low operating tension in R2R process. In low tension, slippage between a moving substrate and a roller would be generated when the operating speed is increased. And scratch would be occurred when slippage is generated due to an air entrainment, which is related to the thickness of the air film. The thickness of the air film is increased when operating speed is increased as shown by simulation based on mathematical model. The occurrence of scratch in conductive pattern printed by roll to roll process is a critical damage because it causes degradation or failure of electrical property of it.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.6
• /
• pp.133-141
• /
• 2008

As interactive multimedia have come into wide use, user interfaces such as remote controllers or classical computer mice have several limitations that cause inconvenience. We propose a vision-based pointing device to resolve this problem. We analyzed the moving image from the camera which is embedded in the pointing device and estimate the movement of the device. The pose of the cursor can be determined from this result. To process in the real time, we used the low resolution of $288{\times}208$ pixel camera and comer points of the screen were tracked using local optical flow method. The distance from screen and device was calculated from the size of screen in the image. The proposed device has simple configurations, low cost, easy use, and intuitive handhold operation like traditional mice. Moreover it shows reliable performance even in the dark condition.

• Proceedings of the KSR Conference
• /
• 2006.11b
• /
• pp.1454-1460
• /
• 2006

This paper assesses communication performance using a roaming simulator when roaming occurs between onboard and ground wireless communication devices for communication based train control system (CBTC). Generally, CBTC is defined as the system regularly collecting location and speed data from each train, transmitting distance information to a train, and optimizing train speed according to train performance. When a train is moving, roaming is also performed to continuously transmit and receive train control information between the ground controller and the train. To operate CBTC, packet loss rate should be less than 1%, roaming time less than 100ms during roaming. We developed a roaming simulator to check communication performance before installing ground and onboard equipments on actual wireless sections. The roaming simulator to be introduced in this paper is for roaming simulation before conducting CBTC field test, which is the project to develop Urban Rail Signaling System Standards, being conducted in KRRI. The simulation consists of one onboard wireless communication device and three ground wireless communication devices, and the roaming simulator estimate packet loss rate occurring during roaming process of the two devices. Therefore, if you use the roaming simulator before the field test, you can predict various problems to occur in actual environment and reduce time, cost and people necessary to resolve these problems.

• Journal of Environmental Science International
• /
• v.19 no.10
• /
• pp.1187-1201
• /
• 2010

The formation and development conditions of the cloud streets over the yellow sea by the Cold Surge of Siberian Anticyclone Expansion which produce the heavy snowfall events over the southwestern coast, Honam District of the Korean peninsula, has been investigated through analyses of the three dimensional snow cloud structures by using the CAPPI, RHI, VAD and VVP data of X-band Radar at Muan Weather Observatory and S-band Radar at Jindo Weather Station. The data to be used are obtained from January 04, 2003, when heavy snow storm hits on Gwangju and Honam District. The PPI Radar images show that the cloud bands distribute in perpendicular to the expansion direction of the high pressure and that the radius of cloud cells is about 5~8 km with 20~30 dBz and distance between each cell is about 10 km. And but the vertical Radar images show that the cloud street is a small scale convective type cloud within height of about 3 km where a stable layer exists. From the VVP images, the time period of the high pressure expansion, the moving direction and development stages of the system are delineated. Finally, the vertical distribution of wind direction is fairly constants, while the wind speed sheer increases with altitude to 3 km.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.1
• /
• pp.226-232
• /
• 2015

As the age of smart device has come, recently many application services related to smart phone are developing. The LBS(Location Based Service) technique is considered as one of the most important techniques to support location based application services. Usually the smart phone acquires the information of position by using the position recognition systems and sensors such as GPS(Global Positioning System) and G-Sensor. However, since the GPS signal from the satellite can hardly be received in the indoor environments, new LBS techniques for the indoor environment are required. In this paper, to solve the problem a position information transceiver using LED lights and smart phone camera sensor is proposed. We proved the possibility of the proposed positioning system through the experiments in the laboratory for the practical verification.