• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.2
• /
• pp.15-23
• /
• 2017

The person Re-identification is the most challenging part of computer vision due to the significant changes in human pose and background clutter with occlusions. The picture from non-overlapping cameras enhance the difficulty to distinguish some person from the other. To reach a better performance match, most methods use feature selection and distance metrics separately to get discriminative representations and proper distance to describe the similarity between person and kind of ignoring some significant features. This situation has encouraged us to consider a novel method to deal with this problem. In this paper, we proposed an enhanced recurrent neural network with three-tier hierarchical network for person re-identification. Specifically, the proposed recurrent neural network (RNN) model contain an iterative expectation maximum (EM) algorithm and three-tier Hierarchical network to jointly learn both the discriminative features and metrics distance. The iterative EM algorithm can fully use of the feature extraction ability of convolutional neural network (CNN) which is in series before the RNN. By unsupervised learning, the EM framework can change the labels of the patches and train larger datasets. Through the three-tier hierarchical network, the convolutional neural network, recurrent network and pooling layer can jointly be a feature extractor to better train the network. The experimental result shows that comparing with other researchers' approaches in this field, this method also can get a competitive accuracy. The influence of different component of this method will be analyzed and evaluated in the future research.

• Journal of the Korean Society for Railway
• /
• v.18 no.2
• /
• pp.139-148
• /
• 2015

By increasing the vertical stiffness of the rail fastening system, the dynamic wheel load of the vehicle can be increased on the ballast track, though this increases the cost of track maintenance. On the other hand, the resistance acting on the wheel is decreased, which lowers the cost of the electric power to run the train. For this reason, the determination of the optimal fastener stiffness is important when attempting to minimize the economic costs associated with both track maintenance and energy to operate the train. In this study, a numerical method for evaluating the optimal vertical stiffness of the fasteners used on ballast track is presented on the basis of the process proposed by L$\acute{o}$pez-Pita et al. They used an approximation formula while calculating the dynamic wheel load. The evaluated fastener stiffness is mainly affected by the calculated dynamic wheel load. In this study, the dynamic wheel load is more precisely evaluated with an advanced vehicle-track interaction model. An appropriate range of the stiffness of the fastener applicable to the design of ballast track along domestic high-speed lines is proposed.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.3
• /
• pp.17-26
• /
• 2003

Recently, as the internet is widespread rapidly among the public, people can use a variety of useful information services through the internet. Accordingly, the protection of information supplied by computer networks 5 has become a matter of primary concern on the whole world. To accede to the realistic demands, it has been worked out some countermeasures to cultivate the experts in information security by the government and many educational facilities. Already the government authority has carried out the each kinds of concerning projects under the framed a policy, Five-Year Development Plan for Information Security Technology. Also, many domestic universities perceives such an international trend, and so they frame their plans to train for the experts in this field, including to found a department with respect to the information security. They are ready to execute their tangible works, such as establishment of educational goal, development of teaching materials, planning curriculum, construction of laboratories and ensuring instructors. Moreover, such universities lead to their students who want to be information security experts to get the fundamental knowledge to lay the foundation for acquiring the information security technology in their bachelor course. In this note, we survey and analyze the curricula of newly-established or member-extended departments with respect to information security fields of some leading universities in the inside and outside of the country, and in conclusion, we propose the effective model of curriculum and educational goal to train the students for the information security experts.

• Journal of Science Education
• /
• v.45 no.1
• /
• pp.90-104
• /
• 2021

This study conducted an online survey to understand what elementary school teachers think about the learning contents of elementary science education subjects needed to train elementary science teachers suitable for the era of the 4th Industrial Revolution. The results are as follows: First, there were many elementary school teachers who thought that the current learning content of elementary science education was not suitable for the era of the 4th Industrial Revolution and that it needed to modify the learning content. Many of the teachers said that the learning content of the subject did not include the characteristics of the 4th Industrial Revolution, but also did not reflect the changes of the times and remained in the past. Second, the content that elementary school teachers thought was important in training elementary school teachers suitable for the era of the 4th Industrial Revolution was mainly related to the interests and curiosity of students, and scientific experiments or inquiry. On the contrary, the items that they thought should be deleted or reduced included science learning theory, science teaching/learning model, nature of science, and guidance for gifted children. Third, the contents that elementary school teachers thought needed to be added as learning content of elementary science education subjects were SSI education, science education-related social change and future prediction, advanced science technology, STEAM guidance, and integrated education within the science field. Fourth, in order to train elementary school teachers suitable for the era of the 4th Industrial Revolution, the contents that they thought should be introduced first as learning content of elementary science education subjects were SSI education, integrated education within the science field, STEAM guidance, and core competencies. Other contents that need to be introduced were software education, safety education, and project learning methods.

• Journal of Industrial Convergence
• /
• v.21 no.1
• /
• pp.159-165
• /
• 2023

While the frequency of seismic occurrence has been increasing recently, the domestic seismic response system is weak, the objective of this research is to compare and analyze the seismic vulnerability of buildings using statistical analysis and machine learning techniques. As the result of using statistical technique, the prediction accuracy of the developed model through the optimal scaling method showed about 87%. As the result of using machine learning technique, because the accuracy of Random Forest method is 94% in case of Train Set, 76.7% in case of Test Set, which is the highest accuracy among the 4 analyzed methods, Random Forest method was finally chosen. Therefore, Random Forest method was derived as the final machine learning technique. Accordingly, the statistical analysis technique showed higher accuracy of about 87%, whereas the machine learning technique showed the accuracy of about 76.7%. As the final result, among the 22,296 analyzed building data, the seismic vulnerabilities of 1,627(0.1%) buildings are expected as more dangerous when the statistical analysis technique is used, 10,146(49%) buildings showed the same rate, and the remaining 10,523(50%) buildings are expected as more dangerous when the machine learning technique is used. As the comparison of the results of using advanced machine learning techniques in addition to the existing statistical analysis techniques, in spatial analysis decisions, it is hoped that this research results help to prepare more reliable seismic countermeasures.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.39-57
• /
• 2012

Recently, due to the introduction of high-tech equipment in interactive exhibits, many people's attention has been concentrated on Interactive exhibits that can double the exhibition effect through the interaction with the audience. In addition, it is also possible to measure a variety of audience reaction in the interactive exhibition. Among various audience reactions, this research uses the change of the facial features that can be collected in an interactive exhibition space. This research develops an artificial neural network-based prediction model to predict the response of the audience by measuring the change of the facial features when the audience is given stimulation from the non-excited state. To present the emotion state of the audience, this research uses a Valence-Arousal model. So, this research suggests an overall framework composed of the following six steps. The first step is a step of collecting data for modeling. The data was collected from people participated in the 2012 Seoul DMC Culture Open, and the collected data was used for the experiments. The second step extracts 64 facial features from the collected data and compensates the facial feature values. The third step generates independent and dependent variables of an artificial neural network model. The fourth step extracts the independent variable that affects the dependent variable using the statistical technique. The fifth step builds an artificial neural network model and performs a learning process using train set and test set. Finally the last sixth step is to validate the prediction performance of artificial neural network model using the validation data set. The proposed model is compared with statistical predictive model to see whether it had better performance or not. As a result, although the data set in this experiment had much noise, the proposed model showed better results when the model was compared with multiple regression analysis model. If the prediction model of audience reaction was used in the real exhibition, it will be able to provide countermeasures and services appropriate to the audience's reaction viewing the exhibits. Specifically, if the arousal of audience about Exhibits is low, Action to increase arousal of the audience will be taken. For instance, we recommend the audience another preferred contents or using a light or sound to focus on these exhibits. In other words, when planning future exhibitions, planning the exhibition to satisfy various audience preferences would be possible. And it is expected to foster a personalized environment to concentrate on the exhibits. But, the proposed model in this research still shows the low prediction accuracy. The cause is in some parts as follows : First, the data covers diverse visitors of real exhibitions, so it was difficult to control the optimized experimental environment. So, the collected data has much noise, and it would results a lower accuracy. In further research, the data collection will be conducted in a more optimized experimental environment. The further research to increase the accuracy of the predictions of the model will be conducted. Second, using changes of facial expression only is thought to be not enough to extract audience emotions. If facial expression is combined with other responses, such as the sound, audience behavior, it would result a better result.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.435-444
• /
• 2019

Recently, DL (Deep Learning) has been rapidly applied in various fields. In particular, classification and object recognition from images are major tasks in computer vision. Most of the DL utilizing imagery is primarily based on the CNN (Convolutional Neural Network) and improving performance of the DL model is main issue. While most CNNs are involve with images for training data, this paper aims to classify and recognize objects using DSM (Digital Surface Model), and slope and aspect information derived from the DSM instead of images. The DSM data sets used in the experiment were established by DGPF (German Society for Photogrammetry, Remote Sensing and Geoinformatics) and provided by ISPRS (International Society for Photogrammetry and Remote Sensing). The CNN-based SegNet model, that is evaluated as having excellent efficiency and performance, was used to train the data sets. In addition, this paper proposed a scheme for training data generation efficiently from the limited number of data. The results demonstrated DSM and derived data could be feasible for semantic classification with desirable accuracy using DL.

• Korean Journal of Acupuncture
• /
• v.21 no.3
• /
• pp.59-76
• /
• 2004

Objectives : The usage of acupuncture has gained popularity for certain chronic pain conditions. However, the efficacy of acupuncture in various diseases has not been fully established and the underlying mechanism is not clearly understood. In the present study, the effect of electroacupuncture (EA) applied to yangno$(SI_6)$ on the neuropathic pain was examined. Methods : A common source of persistent pain in human is a neuropathic pain. Neuropathic pain was induced by tight ligation of L5 spinal nerve. When rats developed pain behaviors, EA was applied for 30 min. under enflurane anesthesia with repeated train stimuli at the intensity of 10X of muscle twitch threshold. The foot withdraw latency of the hind limb was measured for an indicator of pain level after each manipulation. Results : EA increased the mechanical threshold of the foot in the rat model of neuropathic pain significantly for the duration of 1 hr. suggesting a partial alleviation of pain. EA applied to SI6 point produced a significant improvement of mechanical sensitivity of the foot lasting for at least 1 h. However, $ST_{36}$ point did not produce any significant increase of mechanical sensitivity. The improvement of mechanical threshold was interpreted as an analgesic effect. The analgesic effort was specific to the acupuncture point since the analgesic effect on the neuropathic pain model could not be mimicked by EA applied to a point, $ST_{36}$. In addition, this analgesic effect of EA is mediated by a adrenergic mechanism of descending control of spinal cord from the brain. Conclusions : The data suggest that EA produces a potent analgesic effect on the neuropathic pain model in the rat; and 2) that EA-induced analgesia is mediated by a adrenergic mechanism of descending control in a point specific manner.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.139-147
• /
• 2021

This paper introduces and implements a Sound Event Detection (SED) system based on weakly-supervised learning where only part of the data is labeled, and analyzes the effect of parameters. The SED system estimates the classes and onset/offset times of events in the acoustic signal. In order to train the model, all information on the event class and onset/offset times must be provided. Unfortunately, the onset/offset times are hard to be labeled exactly. Therefore, in the weakly-supervised task, the SED model is trained by "strongly labeled data" including the event class and activations, "weakly labeled data" including the event class, and "unlabeled data" without any label. Recently, the SED systems using the mean-teacher model are widely used for the task with several parameters. These parameters should be chosen carefully because they may affect the performance. In this paper, performance analysis was performed on parameters, such as the feature, moving average parameter, weight of the consistency cost function, ramp-up length, and maximum learning rate, using the data of DCASE 2020 Task 4. Effects and the optimal values of the parameters were discussed.

• Journal of Advanced Navigation Technology
• /
• v.24 no.6
• /
• pp.620-626
• /
• 2020

This research propose and simulate a solar power generation system monitoring system based on Modbus TCP communication using RaspberryPi, an IOT equipment, as a master and an inverter as a slave. In this model, various sensors are added to the RaspberryPi to add necessary information for monitoring solar power plants, and power generation prediction and monitoring information are transmitted to the smart phone through real-time power generation prediction. In addition, information that is continuously generated by the solar power plant is built on the server as big data, and a deep learning model for predicting power generation is trained and updated. As a result of the study, stable communication was possible based on Modbus TCP with the Raspberry Pi in the inverter, and real-time prediction was possible with the deep learning model learned in the Raspberry Pi. The server was able to train various deep learning models with big data, and it was confirmed that LSTM showed the best error with a learning error of 0.0069, a test error of 0.0075, and an RMSE of 0.0866. This model suggested that it is possible to implement a real-time monitoring system that is simpler, more convenient, and can predict the amount of power generation for inverters of various manufacturers.