• Proceedings of the KSPS conference
• /
• 2006.11a
• /
• pp.142-145
• /
• 2006

Discrimination between speech and music is important in many multimedia applications. Previously we proposed a new parameter for speech/music discrimination, the mean of minimum cepstral distances (MMCD), and it outperformed the conventional parameters. One weakness of it is that its performance depends on range of candidate frames to compute the minimum cepstral distance, which requires the optimal selection of the range experimentally. In this paper, to alleviate the problem, we propose a multi-dimensional MMCD parameter which consists of multiple MMCDs with different ranges of candidate frames. Experimental results show that the multi-dimensional MMCD parameter yields an error rate reduction of 22.5% compared with the optimally chosen one-dimensional MMCD parameter.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.18 no.3
• /
• pp.234-243
• /
• 2015

It is required to have the capability to predict the impact point of the ballistic target in order to assign the firing unit with high engagement possibility for the interception in the ballistic target defense systems. In this paper, a novel method is proposed to predict the impact point of the ballistic target using a flight phase discrimination algorithm given the insufficient measurements on the partial trajectory. The flight of a ballistic target is composed of a boost phase and a ballistic phase with different dynamics. The flight phase is discriminated by using the normalized innovation distance between measurements and a priori estimated measurements. The threshold and tolerance in the flight phase discrimination are determined from the probabilistic characteristics of the estimation error. Monte Carlo simulations are performed to verify the proposed method.

• Journal of Biosystems Engineering
• /
• v.24 no.3
• /
• pp.225-234
• /
• 1999

This study was performed as a basic research for establishing an objective quality evaluation method on whole bulbs of garlic. The size of a whole bulb of garlic, the number and the uniformity of complete individual garlics, and the existence of bad individual garlics in the whole bulb of garlic were selected as quality grading factors. Quality discrimination algorithms with machine vision techniques were developed and verified for the four factors based on morphological and physical features of whole bulbs of garlic. Based on the results, the size discrimination by the projected area of a whole bulbs of garlic suggested four grading levels and the algorithm for predicting the number of complete individual garlics based on the peaks on its projected boundary showed ${\pm}$0.78 prediction error. In addition, the uniformity represented by coefficient of variation could be divided into four levels, but the algorithm for discriminating the existence of bad individual garlics in a whole bulb of garlic was not effective.

• Journal of electromagnetic engineering and science
• /
• v.7 no.2
• /
• pp.83-90
• /
• 2007

In this paper, the formulation of the protection ratio based upon a composite fade margin and availability is newly presented for the detailed planning of frequency coordination in the microwave relay system network, and computed results for co-channel and adjacent channel protection ratios are illustrated over an actual system with 6.2 GHz. It is shown that the protection ratio to assure a quality of service can be expressed in terms of the composite fade margin, noise-to-interference ratio, net filter discrimination, and system parameters. In addition, the net filter discrimination, depending upon the transmitter spectrum mask and the overall receiver filter characteristic, has been examined to investigate the effect of the adjacent channel protection ratio caused by the adjacent channel interference. Regarding simulated results for 6.2 GHz, 60 km, 64-QAM, and N/I=6 dB at the bit error rate of $10^{-6}$, composite fade margin and co-channel protection ratio yield 25.14 and 50.3 dB, respectively. Also, the net filter discrimination of 26.5 dB and the adjacent channel protection ratio of 23.8 dB are obtained at the first adjacent channel of 30 MHz. The proposed method provides some merits in view of a comprehensive and practical application with more detailed and various system parameters needed to access the criteria for making the proper frequency coordination.

• Nuclear Engineering and Technology
• /
• v.51 no.5
• /
• pp.1355-1364
• /
• 2019

A recently published nuclear forensics methodology for source discrimination of separated weapons-grade plutonium utilizes intra-element isotope ratios and a maximum likelihood formulation to identify the most likely source reactor-type, fuel burnup and time since irradiation of unknown material. Sensitivity studies performed here on the effects of random measurement error and the uncertainty in intra-element isotope ratio values show that different intra-element isotope ratios have disproportionate contributions to the determination of the reactor parameters. The methodology is robust to individual errors in measured intra-element isotope ratio values and even more so for uniform systematic errors due to competing effects on the predictions from the selected intra-element isotope ratios suite. For a unique sample-model pair, simulation uncertainties of up to 28% are acceptable without impeding successful source-reactor discrimination. However, for a generic sample with multiple plausible sources within the reactor library, uncertainties of 7% or less may be required. The results confirm the critical role of accurate reactor core physics, fuel burnup simulations and experimental measurements in the proposed methodology where increased simulation uncertainty is found to significantly affect the capability to discriminate between the reactors in the library.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.05a
• /
• pp.1-5
• /
• 2002

We consider the problem of optimal bandwidth choice for nonparametric classification, based on kernel density estimators, where the problem of interest is distinguishing between two univariate distributions. When the densities intersect at a single point, optimal bandwidth choice depends on curvatures of the densities at that point. The problem of empirical bandwidth selection and classifying data in the tails of a distribution are also addressed.

• Korean Journal of Food Science and Technology
• /
• v.38 no.6
• /
• pp.835-837
• /
• 2006

The geographical origin of Tricholoma matsutake mushrooms was studied using near-infrared spectroscopy. Modified partial least-square regression analyses were used to discriminate geographical origin. Two-hundred fifty-six of 259 actual domestic Tricholoma matsutake were classified as domestic produce, Sixty of 81 actual imported mushrooms were correctly classified as imported, while the other 21 imported from North Korea were not clearly classified. The accuracy of geographical origin discrimination was 92.94% The correlation coefficient, standard error of calibration, and standard error of prediction from modified partial least-square regression analysis were 0.84, 15.10% and 18.30% respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.3C
• /
• pp.125-132
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Futhermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about 5°horizontally and 8°vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.3 no.1
• /
• pp.38-46
• /
• 2010

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Furthermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about $5^{\circ}$horizontally and $8^{\circ}$vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.2
• /
• pp.322-329
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNS). With GRNNS, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. furthermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10$\%$ improvement in classification error. The angular gaze accuracy is about $5^{circ}$horizontally and $8^{circ}$vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.