• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4117-4135
• /
• 2020

Due to the angle of view, illumination and scene diversity, real-time dynamic face detection and recognition is no small difficulty in those unrestricted environments. In this study, we used the intrinsic correlation between detection and calibration, using a multi-task cascaded convolutional neural network(MTCNN) to improve the efficiency of face recognition, and the output of each core network is mapped in parallel to a compact Euclidean space, where distance represents the similarity of facial features, so that the target face can be identified as quickly as possible, without waiting for all network iteration calculations to complete the recognition results. And after the angle of the target face and the illumination change, the correlation between the recognition results can be well obtained. In the actual application scenario, we use a multi-camera real-time monitoring system to perform face matching and recognition using successive frames acquired from different angles. The effectiveness of the method was verified by several real-time monitoring experiments, and good results were obtained.

• Journal of electromagnetic engineering and science
• /
• v.13 no.4
• /
• pp.233-239
• /
• 2013

This paper proposes an accurate approach for predicting transferred power from a noise source to integrated circuits based on the characteristics of the power transfer network. A power delivery trace on a package and a printed circuit board are designed to transmit power from an external source to integrated circuits. The power is demonstrated between an injection terminal on the edge of the printed circuit board and integrated circuits, and the power transfer function of the power distribution network is derived. A two-tier calibration is applied to the test, and scattering parameters of the network are measured for the calculation of the power transfer function. After testing to obtain the indispensable parameters, the real received and tolerable power of the integrated circuits can be easily achieved. Our proposed estimation method is an enhancement of the existing the International Electrotechnical Commission standard for precise prediction of the electromagnetic immunity of integrated circuits.

• Journal of Sensor Science and Technology
• /
• v.28 no.5
• /
• pp.284-288
• /
• 2019

A calibration-free analog neuron circuit is proposed as a viable alternative to the power hungry digital neuron in implementing a deep neural network. The conventional analog neuron requires calibrations because a voltage-mode link is used between the soma and the synapse, which results in significant uncertainty in terms of current mapping. In this work, a current-mode link is used to establish a robust link between the soma and the synapse against the variations in the process and interconnection impedances. The increased hardware owing to the adoption of the current-mode link is estimated to be manageable because the number of neurons in each layer of the neural network is typically bounded. To demonstrate the utility of the proposed analog neuron, a simple neural network with $4{\times}7{\times}3$ architecture has been designed for classifying iris flowers. The chip is now under fabrication in 0.35 mm CMOS technology. Thus, the proposed true current-mode analog neuron can be a practical option in realizing power-efficient neural networks for edge computing.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.04a
• /
• pp.733-736
• /
• 1996

Neural Network(NN) is known to be suitable for forecasting corporate bankruptcy because of discriminant capability. Bankruptcy prediciton on NN by now has mostly been studied based on financial indices at specific point of time. However, the financial profile of corporates fluctuates within a certain range with the elapse of time. Besides, we need a lot of data of different bankrupt types in order to apply NN for better bankruptcy prediciton. Therefore, we have decided to focus on textiles and clothing industries for bankruptcy prediction with limited data. One part of the collected data was used for training and calibration, and the other was used for verification. The model makes a learning with extended data from financial indices at specific point of time. The trained model has been tested and we could get a high hitting ratio relatively.

• Korean Management Science Review
• /
• v.14 no.2
• /
• pp.91-111
• /
• 1997

Neural Network(NN) is known to be suitable for forecasting corporate bankruptcy because of discriminant capability. Bandkruptcy prediction on NN by now has mostly been studied based on financial indices at specific point of time. However, the financial profile of corporates fluctuates within a certain range with the elapse of time. Besides, we need a lot of data of different bankrupt types in order to apply NN for better bankruptcy prediction. Therefore, We have decided to focus on textile and clothing industries for bankruptcy prediction with limited data. One part of the collected data was used for training and calibration, and the other was used for verification. The model makes a learning with extended data from financial indices at specific point of time. The trained model has been tested and we could get a high hitting ratio relatively.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.5
• /
• pp.39-44
• /
• 1998

Determining the absolute location of a mobile robot is essential in the navigation of a mobile robot. In this paper, a method to determine the position of a mobile robot through the visual image of a landrnark using neural networks is proposed. In determining the position of a mobile robot on the world coordinate, there is a position error because of uncertainty in pixels, incorrect camera calibration and lens distortion. To reduce the errors, a method using a BPNN(Back Propagation Neural Network) is proposed. The experimental results are presented to illustrate the superiority of the proposed method when comparing with the conventional methods.

• Journal of Advanced Marine Engineering and Technology
• /
• v.27 no.7
• /
• pp.921-929
• /
• 2003

An experimental investigation was performed to study the characteristics of turbulent water flow in a horizontal circular tube by using liquid crystal. To determine some characteristics of the turbulent flow, 2D PIV technique is employed for velocity measurement and liquid crystal is used for heat transfer experiments in water. Temperature visualization was made quantitatively by calibrating the color of the liquid crystal versus temperature using various approaches (TLC technique: Thermochromic Liquid Crystal), and a neural-network algorithm was applied to the color-to-temperature calibration. This study shoud the temperature and time-mean velocity distribution for Re = 2,436, 2,500 and 2,724 along longitudinal sections and the results appear to be physically reasonable.

• Journal of Power System Engineering
• /
• v.20 no.6
• /
• pp.51-57
• /
• 2016

Experiment is carried out to investigate the mixed convective flow in three-dimensional horizontal rectangular channels filled with high viscous fluid. The particle image velocimetry(PIV) with thermo-sensitive liquid crystal tracers is used for visualizing and analysis. Quantitative data of temperature and velocity are obtained by applying the color-image processing to a visualized image, and neural network is applied to the color-to-temperature calibration. In this study, the fluid used is silicon oil(Pr=909), the aspect ratio(channel width to heigh) is 4 and Reynolds number is $2{\times}10^{-2}$. From the present study, we can visualize the quantitative temperature and velocity of mixed convective flow in three-dimensional horizontal rectangular channels simultaneously.

• Journal of Korean Society of Water and Wastewater
• /
• v.13 no.2
• /
• pp.23-33
• /
• 1999

This study is to investigate the effect of C factor errors on the analysis of water distribution systems. For this purpose, an artificial distribution network and a real distribution network were selected as the study networks. Results are as follows. 1. The C factor of a pipe which has small velocity didn't give significant effect on the analysis of a water distribution system. 2. The effect of decreased value of C factors give more influence on the analysis of water distribution systems than that of the increased values. 3. For the C factor calibration, errors of the residual water heads as well as those of the head losses should be considered together. 4. In the analysis of water distribution systems, changes of C factors can give influences only on the nodes which locate behind the pipe. Therefore, this characteristics should be considered in the selection of nodes for the measurement of water heads.

• Proceedings of the IEEK Conference
• /
• 2001.06a
• /
• pp.169-172
• /
• 2001

This paper has been studied a calibration techniques for monopole antenna in the frequency range 150 KHz to 30 MHz. The long wavelength associated with the low frequency, methods used to calibrate or characterize antennas at higher frequencies are not applicable. The equivalent capacitance substitution method uses a dummy antenna in place of the actual rod element See figure 1. for guidance in making a dummy antenna. Set up the matching network to be characterized and the measuring equipment as shown in Figure 2. Subtract the measured output of the matching network from the measured output of the signal generator and subtract -6 dB(for the 1 m rod). Measurements made at a sufficient number of frequencies number of frequencies to obtain a smooth curve of antenna factor.(fig 5.)