• The Journal of the Acoustical Society of Korea
• /
• v.28 no.2
• /
• pp.112-118
• /
• 2009

This paper proposes a rapid detection algorithm of a salient region for music video browsing system, which can be applied to mobile device and digital video recorder (DVR). The input music video is decomposed into the music and video tracks. For the music track, the music highlight including musical chorus is detected based on structure analysis using energy-based peak position detection. Using the emotional models generated by SVM-AdaBoost learning algorithm, the music signal of the music videos is classified into one of the predefined emotional classes of the music automatically. For the video track, the face scene including the singer or actor/actress is detected based on a boosted cascade of simple features. Finally, the salient region is generated based on the alignment of boundaries of the music highlight and the visual face scene. First, the users select their favorite music videos from various music videos in the mobile devices or DVR with the information of a music video's emotion and thereafter they can browse the salient region with a length of 30-seconds using the proposed algorithm quickly. A mean opinion score (MOS) test with a database of 200 music videos is conducted to compare the detected salient region with the predefined manual part. The MOS test results show that the detected salient region using the proposed method performed much better than the predefined manual part without audiovisual processing.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• Journal of KIISE:Software and Applications
• /
• v.33 no.12
• /
• pp.1073-1089
• /
• 2006

The goal of Natural Language Processing(NLP) is to make a computer understand a natural language and to deliver the meanings of natural language to humans. Word sense Disambiguation(WSD is a very important technology to achieve the goal of NLP. In this paper, we describe a technology for automatic homonyms disambiguation using both Mutual Information(MI) and a Sense-Tagged Compound Noun Dictionary. Previous research work using word definitions in dictionary suffered from the problem of data sparseness because of the use of exact word matching. Our work overcomes this problem by using MI which is an association measure between words. To reflect language features, the rate of word-pairs with MI values, sense frequency and site of word definitions are used as weights in our system. We constructed a Sense-Tagged Compound Noun Dictionary for high frequency compound nouns and used it to resolve homonym sense disambiguation. Experimental data for testing and evaluating our system is constructed from QA(Question Answering) test data which consisted of about 200 query sentences and answer paragraphs. We performed 4 types of experiments. In case of being used only MI, the result of experiment showed a precision of 65.06%. When we used the weighted values, we achieved a precision of 85.35% and when we used the Sense-Tagged Compound Noun Dictionary, we achieved a precision of 88.82%, respectively.

• Journal of the Korean Society of Radiology
• /
• v.10 no.2
• /
• pp.81-87
• /
• 2016

In this study we proposed a texture feature analysis algorithm that distinguishes between a normal image and a diseased image using CT images of some fatty liver patients, and generates both Eigen images and test images which can be applied to the proposed computer aided diagnosis system in order to perform a quantitative analysis for 6 parameters. And through the analysis, we derived and evaluated the recognition rate of CT images of fatty liver. As the results of examining over 30 example CT images of fatty liver, the recognition rates representing a specific texture feature-value are as follows: some appeared to be as high as 100% including Average Gray Level, Entropy 96.67%, Skewness 93.33%, and Smoothness while others showed a little low disease recognition rate: 83.33% for Uniformity 86.67% and for Average Contrast 80%. Consequently, based on this research result, if a software that enables a computer aided diagnosis system for medical images is developed, it will lead to the availability for the automatic detection of a diseased spot in CT images of fatty liver and quantitative analysis. And they can be used as computer aided diagnosis data, resulting in the increased accuracy and the shortened time in the stage of final reading.

• Science of Emotion and Sensibility
• /
• v.23 no.2
• /
• pp.3-12
• /
• 2020

The purpose of this work is to suggest the optimal color temperature, which induces a sense of comfort for autonomous vehicle users through the analysis of biosignal using electroencephalography (EEG) and photoplethysmography (PPG). To achieve this purpose, we applied lighting with a color temperature of 3000 K, 4000 K, 5000 K, and 6000 K to the autonomous driving environment. We experimented in a laboratory equipped with a graphic driving simulator. The experimental procedure is as follows: 1) stabilization (5 min). 2) Uchida-Kraepelin test (3 min). 3) Automatic driving + lighting (3 min). This procedure was repeated four times under different color temperatures. We performed frequency analysis on a collected time-series data and calculated the power value for each frequency band through power spectrum analysis. In the case of EEG, we analyzed α- and β-waves, which are indicators of stability and arousal, respectively. In the case of PPG, we analyzed the sympathetic nervous system activity. To reduce deviations between the subjects, we normalized the data before analysis. The result of the first analysis revealed that α-wave increased only at 5000 K, while the β-wave increased at almost all color temperatures. In addition, in the case of PPG, sympathetic nervous system activity (SNSA) increased under driving conditions. The result of the second analysis revealed that the difference between β-wave and SNSA is insignificant. In conclusion, the increase in α-waves showed that EEG was most stable at 5000 K. The results of this study can be applied to the upcoming autonomous driving era to induce high driver satisfaction. Furthermore, this approach could eventually lead to the acceptance of autonomous vehicles by suggesting a positive effect of autonomous driving.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.5
• /
• pp.8-14
• /
• 2009

In the research field of speech recognition, pinpointing the endpoints of speech utterance even with the presence of background noise is of great importance. These noise present during recording introduce disturbances which complicates matters since what we just want is to get the stationary parameters corresponding to each speech section. One major cause of error in automatic recognition of isolated words is the inaccurate detection of the beginning and end boundaries of the test and reference templates, thus the necessity to find an effective method in removing the unnecessary regions of a speech signal. The conventional methods for speech endpoint detection are based on two linear time-domain measurements: the short-time energy, and short-time zero-crossing rate. They perform well for clean speech but their precision is not guaranteed if there is noise present, since the high energy and zero-crossing rate of the noise is mistaken as a part of the speech uttered. This paper proposes a novel approach in finding an apparent threshold between noise and speech based on Lyapunov Exponents (LEs). This proposed method adopts the nonlinear features to analyze the chaos characteristics of the speech signal instead of depending on the unreliable factor-energy. The excellent performance of this approach compared with the conventional methods lies in the fact that it detects the endpoints as a nonlinearity of speech signal, which we believe is an important characteristic and has been neglected by the conventional methods. The proposed method extracts the features based only on the time-domain waveform of the speech signal illustrating its low complexity. Simulations done showed the effective performance of the Proposed method in a noisy environment with an average recognition rate of up 92.85% for unspecified person.

• Journal of Bio-Environment Control
• /
• v.26 no.3
• /
• pp.215-220
• /
• 2017

This study was conducted to improve the insufficiency of fruit vegetable grafting system developed by National Institute of Agricultural Sciences, Rural Development Administration. When the rotary blade cut the stem of scions and rootstocks, the grafting failure at curved cutting surfaces happened. The cutting depth of a tomato seedling by a rotated cutter was calculated 0.11 mm even when the cutting arm length and the maximum stem diameter were 50 mm and 5 mm, respectively. Mathematical analysis and high-speed photography showed that there was no problem by cutting in straight the stem of scions and rootstocks. The compression test of seedling stems to design the optimal shape of gripper showed that stems were not completely restored when they were compressed above 0.8 mm and 0.6 mm in case of rootstocks and scion, respectively. This study found that the bending angle of stem of tomato seedlings at the grafting period was 10 degree on average. The optimal gripper finger was the edge finger type which could be precisely set center point by adjusting the distance between fingers. In addition, it was found that most of seedling could be grasped without damage when the finger-to-finger distances is set to 2.5 mm for scion and 3.0 mm for rootstocks and finger are coated by 1 mm-thick flexible material.

• Journal of the Korean Society of Radiology
• /
• v.9 no.6
• /
• pp.369-374
• /
• 2015

In this study we proposed a texture feature analysis algorithm that distinguishes between a normal image and a diseased image using CT images of some brain hemorrhage patients, and generates both Eigen images and test images which can be applied to the proposed computer aided diagnosis system in order to perform a quantitative analysis for 6 parameters. And through the analysis, we derived and evaluated the recognition rate of CT images of brain hemorrhage. As the results of examining over 40 example CT images of brain hemorrhage, the recognition rates representing a specific texture feature-value are as follows: some appeared to be as high as 100% including average gray level, average contrast, smoothness, and Skewness while others showed a little low disease recognition rate: 95% for uniformity and 87.5% for entropy. Consequently, based on this research result, if a software that enables a computer aided diagnosis system for medical images is developed, it will lead to the availability for the automatic detection of a diseased spot in CT images of brain hemorrhage and quantitative analysis. And they can be used as computer aided diagnosis data, resulting in the increased accuracy and the shortened time in the stage of final reading.

• Economic and Environmental Geology
• /
• v.29 no.6
• /
• pp.713-724
• /
• 1996

The time-series resident solute concentrations, monitored at two field plots using the automated 144-channel TDR system by Kim (this issue), are used to investigate the dominant transport mechanism at field scale. Two models, based on contradictory assumptions for describing the solute transport in the vadose zone, are fitted to the measured mean breakthrough curves (BTCs): the deterministic one-dimensional convection-dispersion model (CDE) and the stochastic-convective lognormal transfer function model (CLT). In addition, moment analysis has been performed using the probability density functions (pdfs) of the travel time of resident concentration. Results of moment analysis have shown that the first and second time moments of resident pdf are larger than those of flux pdf. Based on the time moments, expressed in function of model parameters, variance and dispersion of resident solute travel times are derived. The relationship between variance or dispersion of solute travel time and depth has been found to be identical for both the time-series flux and resident concentrations. Based on these relationships, the two models have been tested. However, due to the significant variations of transport properties across depth, the test has led to unreliable results. Consequently, the model performance has been evaluated based on predictability of the time-series resident BTCs at other depths after calibration at the first depth. The evaluation of model predictability has resulted in a clear conclusion that for both experimental sites the CLT model gives more accurate prediction than the CDE model. This suggests that solute transport at natural field soils is more likely governed by a stream tube model concept with correlated flow than a complete mixing model. Poor prediction of CDE model is attributed to the underestimation of solute spreading and thus resulting in an overprediction of peak concentration.

• The KIPS Transactions:PartB
• /
• v.9B no.5
• /
• pp.663-672
• /
• 2002

In this paper, Using an automatic tracking system, behavior of an aquatic insect, Chironomus sp. (Chironomidae), was observed in semi-natural conditions in response to sub-lethal treament of a carbamate insecticide, carbofuran. The fourth instar larvae were placed in an observation cage $(6cm\times{7cm}\times{2.5cm)}$ at temperature of $18^\circ{C}$ and the light condition of 10 time (light) : 14 time (dark). The tracking system was devised to detect the instant, partial movement of the insect body. Individual movement was traced after the treatment of carbofuran (0.1ppm) for four days 2days : before treatment, 2 days : after treatment). Along with the other irregular behaviors, "ventilation activity", appearing as a shape of "compressed zig-zag", was more frequently observed after the treatment of the insecticide. The activity of the test individuals was also generally depressed after the chemical treatment. In order to detect behavioral changes of the treated specimens, wavelet analysis was implemented to characterize different movement patterns. The extracted parameters based on Discrete Wavelet Transforms (DWT) were subsequently provided to artificial neural networks to be trained to represent different patterns of the movement tracks before and after treatments of the insecticide. This combined model of wavelets and artificial neural networks was able to point out the occurrence of characteristic movement patterns, and could be an alternative tool for automatically detecting presences of toxic chemicals for water quality monitoring. quality monitoring.