• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1175-1186
• /
• 2017

Human emotion recognition is one of the promising applications in the era of artificial super intelligence. Thus far, facial expression traits are considered to be the most widely used information cues for realizing automated emotion recognition. This paper proposes a novel facial expression recognition (FER) method that works well for recognizing emotion from image sequences. To this end, we develop the so-called weighted soft voting classification (WSVC) algorithm. In the proposed WSVC, a number of classifiers are first constructed using different and multiple feature representations. In next, multiple classifiers are used for generating the recognition result (namely, soft voting) of each face image within a face sequence, yielding multiple soft voting outputs. Finally, these soft voting outputs are combined through using a weighted combination to decide the emotion class (e.g., anger) of a given face sequence. The weights for combination are effectively determined by measuring the quality of each face image, namely "peak expression intensity" and "frontal-pose degree". To test the proposed WSVC, CK+ FER database was used to perform extensive and comparative experimentations. The feasibility of our WSVC algorithm has been successfully demonstrated by comparing recently developed FER algorithms.

• Proceedings of the KSR Conference
• /
• 2000.11a
• /
• pp.17-23
• /
• 2000

The purpose of this paper is to evaluate the relative efficiency of Korean National Railroad (KNR). Railways can be seen as multi-product firms which produced both passenger and freight transportation services. For evaluating the operational efficiencies of 23 UIC (International Union of Railways) members including KNR we use data envelopment analysis (DEA) in which multiple inputs and multiple outputs can be handled explicitly.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.18 no.35
• /
• pp.165-175
• /
• 1995

The objective of this study is to evaluate the reliability of DEA for the measurement of efficiency of Commercial banks in Korea using the DEA methods. DEA is basically a mathematical programing technique initially developed by Chanes, Cooper and Rhodes(197n) to evaluate the relative efficiency of not-for-profit organizations where multiple outputs are produced with multiple inputs. This study show that DEA approach is not only usefulness in evaluating performance of banks but also a useful bank management tool.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.41.5-41
• /
• 2001

A novel method is introduced for determining the weights of a regularization network which approximates one-to-many mapping. A conventional neural network will converges to the average value when outputs are multiple for one input. The capability of proposed network is demonstrated by an example of learning inverse mapping.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.11 no.3
• /
• pp.135-145
• /
• 2011

Accurate process characterization and optimization are the first step for a successful advanced process control (APC), and they should be followed by continuous monitoring and control in order to run manufacturing processes most efficiently. In this paper, process characterization and recipe optimization methods with multiple outputs are presented in high density plasma-chemical vapor deposition (HDP-CVD) silicon dioxide deposition process. Five controllable process variables of Top $SiH_4$, Bottom $SiH_4$, $O_2$, Top RF Power, and Bottom RF Power, and two responses of interest, such as deposition rate and uniformity, are simultaneously considered employing both statistical response surface methodology (RSM) and neural networks (NNs) based genetic algorithm (GA). Statistically, two phases of experimental design was performed, and the established statistical models were optimized using performance index (PI). Artificial intelligently, NN process model with two outputs were established, and recipe synthesis was performed employing GA. Statistical RSM offers minimum numbers of experiment to build regression models and response surface models, but the analysis of the data need to satisfy underlying assumption and statistical data analysis capability. NN based-GA does not require any underlying assumption for data modeling; however, the selection of the input data for the model establishment is important for accurate model construction. Both statistical and artificial intelligent methods suggest competitive characterization and optimization results in HDP-CVD $SiO_2$ deposition process, and the NN based-GA method showed 26% uniformity improvement with 36% less $SiH_4$ gas usage yielding 20.8 ${\AA}/sec$ deposition rate.

• Korean Management Science Review
• /
• v.32 no.4
• /
• pp.193-207
• /
• 2015

Government-funded research institutes (GRIs) have played a pivotal role in national R&D in Korea. To achieve desired goals of GRIs with the limited R&D budget, their performance along with time needs to be measured and compared so that appropriate R&D policies can be formulated and implemented. This study measures the dynamic performance of GRIs from the efficiency perspective using the window model of data envelopment analysis (DEA). DEA is a non-parametric approach to measuring the relative efficiency of decision-making units (DMUs) with multiple inputs and outputs, and the DEA window model can capture the dynamic changes in efficiency of DMUs during multiple periods. The relative efficiency of GRIs is measured from the two perspectives: R&D and R&BD. Patents, papers, technology transfers are selected as outputs for R&D while compensated technology transfers and technology royalty are employed as outputs for R&BD. This study measures and compares the two types of performance of 20 Korean GRIs under the control of National Research Council of Science and Technology during the period of six years from 2008 to 2013. The results are expected to provide fruitful implications for national R&D policy making.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.4
• /
• pp.451-455
• /
• 2008

In this paper, the k-nearest templates method is proposed to combine multiple classifiers effectively. First, the method decomposes training samples of each class into several subclasses based on the outputs of classifiers to represent a class as multiple models, and estimates a localized template by averaging the outputs for each subclass. The distances between a test sample and templates are then calculated. Lastly, the test sample is assigned to the class that is most frequently represented among the k most similar templates. In this paper, C-means clustering algorithm is used as the decomposition method, and k is automatically chosen according to the intra-class compactness and inter-class separation of a given data set. Since the proposed method uses multiple models per class and refers to k models rather than matches with the most similar one, it could obtain stable and high accuracy. In this paper, experiments on UCI and ELENA database showed that the proposed method performed better than conventional fusion methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.1C
• /
• pp.24-31
• /
• 2003

In this paper, a new opto-digital multiple information hiding and real-time extracting system is implemented. That is, multiple information is hidden in a cover image by using the stego keys which are generated by combined use of random sequence(RS) and Hadamard matrix(HM) and these hidden information is extracted in real-time by using a new optical correlator-based extraction system. In the experiment, 3 kinds of information, English alphabet of "N", "R", "L" having 512$\times$512 pixels, are formulated 8$\times$8 blocks and each of these information is multiplied with the corresponding stego keys having 64$\times$64 pixels one by one. And then, by adding these modulated data to a cover image of "Lena"having 512$\times$512 pixels, a stego image is finally generated. In this paper, as an extraction system, a new optical nonlinear joint transform correlator(NJTC) is introduced to extract the hidden data from a stego image in real-time, in which optical correlation between the stego image and each of the stego keys is performed and from these correlation outputs the hidden data can be asily exacted in real-time. Especially, it is found that the SNRs of the correlation outputs in the proposed optical NJTC-based extraction system has been improved to 7㏈ on average by comparison with those of the conventional JTC system under the condition of having a nonlinear parameter less than k=0.4. This good experimental results might suggest a possibility of implementation of an opto-digital multiple information hiding and real-time extracting system.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.2
• /
• pp.89-96
• /
• 2017

We develop an embedded system to measure various sensor data, control multiple motors, and communicate with mobile devices for system managements. Choosing TI AM3359 microprocessor featuring high processing performance, low power consumption, and various I/O device support, we design and build the embedded system hardware so that it supports multiple global positioning system (GPS) and gyro sensor modules to measure precise position; multiple pulse width modulation (PWM) outputs to control multiple direct current (DC) motors; a Bluetooth module to communicate with mobile devices. Then, we port the boot loader and device drivers to the built circuit board and construct the firmware development environment for the application programming. The performance of the designed and implemented embedded system is demonstrated by real motor control test using GPS and gyro sensor data and control parameters configured by a mobile device.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.3
• /
• pp.251-258
• /
• 2003

Optical Switching Matrix (OSM) or Optical Multistage Interconnection Networks (OMINs) comprising photonic switches have been studied extensively as important interconnecting blocks for Optical Cross Connects (OXC) based on Optical Burst Switching (OBS). A basic element of photonic switching networks is a 2$\times$2 directional coupler with two inputs and two outputs. This paper is concerned with the diagnosis of multiple crosstalk-faults in OSM. As the network size becomes larger in these days, the conventional diagnosis methods based on tests and simulation become inefficient, or even more impractical. We propose a simple and easily implementable algorithm for detection and isolation of the multiple crosstalk-faults in OSM. Specifically. we develop an algorithm for isolation of the source fault in switching elements whenever the multiple crosstalk-faults arc detected in OSM. The proposed algorithm is illustrated by an example of 16$\times$16 OSM.