• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.7A
• /
• pp.1052-1063
• /
• 1999

In designing the quantizer of a coding scheme using a training sequence (TS), the training algorithm tries to find a quantizer that minimizes the distortion measured in the TS. In order to evaluate the trained quantizer or compare the coding scheme, we can observe the minimized distortion. However, the minimized distortion is a biased estimate of the minimal distortion for the input distribution. Hence, we could often overestimate a quantizer or make a wrong comparison even if we use a validating sequence. In this paper, by compensating the minimized distortion for the TS, a new estimate is proposed. Compensating term is a function of the training ratio, the TS size to the codebook size. Several numerical results are also introduced for the proposed estimate.

• Journal of the Ergonomics Society of Korea
• /
• v.16 no.1
• /
• pp.97-105
• /
• 1997

An aircraft simulator for ergonomics testing and pilot training was developed from the joint work Agency for Defense Development(ADD) and Daewoo Heavy Industry, LTD, in Korea at first time. It is basically to satisfy the requirements established by FAA-AC-120-40C ( 1995-JAN-26). The aircraft simulator will be used mainly for ergonomics testing and pilot training for basic trainer on ADD and Korea Air Force in near futrue. This simulator reproduces faithfully the cockpit and flight characteristics of the KTX-1 aircraft. It is one of the latest full flight simulators that have the CGI(computer graphic image) visual system and six degree of freedom motions system. Development efforts focused on user-oriented design approach for ergonomics testing and flight training of pilots. Main characteristics of each subsystem are described such as cockpit, instruments, control loading system, motion system, visual system, aural system, instructor operation station and aircraft simulation softwear.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.5
• /
• pp.59-67
• /
• 1998

This paper propose a new vector-quantizer design method for coefficients of wavelet transformed images. In conventional wavelet transform, it is quite often to employ wavelet transformed coefficients, not containing images to be encoded, as training sequences for designing a vector-quantizer. This method has a serious drawback ; it is not known how to find a proper set of training images. This paper investigates characteristics of images that should be considered in the design of vector-quantizers for wavelet transformed images. Besides the statistical parameters such as correlation and standard deviation, edge components are shown to characterise wavelet transform images. Training sequences established in accordance with the above knowledge are used in the design of quantizers having guaranteed range of applicable images. Results of computer simulations are shown to demonstrate the effectiveness of the proposed method.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.809-819
• /
• 2020

Gender prediction accuracy increases as convolutional neural network (CNN) architecture evolves. This paper compares voting and ensemble schemes to utilize the already trained five CNN models to further improve gender prediction accuracy. The majority voting usually requires odd-numbered models while the proposed softmax-based voting can utilize any number of models to improve accuracy. The ensemble of CNN models combined with one more fully-connected layer requires further tuning or training of the models combined. With experiments, it is observed that the voting or ensemble of CNN models leads to further improvement of gender prediction accuracy and that especially softmax-based voters always show better gender prediction accuracy than majority voters. Also, compared with softmax-based voters, ensemble models show a slightly better or similar accuracy with added training of the combined CNN models. Softmax-based voting can be a fast and efficient way to get better accuracy without further training since the selection of the top accuracy models among available CNN pre-trained models usually leads to similar accuracy to that of the corresponding ensemble models.

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1242-1250
• /
• 2021

Detection of outliers deviating normal data distribution in high dimensional data is an important technique in many application areas. In this paper, a distance-based outlier detection method using landmarks in high dimensional data is proposed. Given normal training data, the k-means clustering method is applied for the training data in order to extract the centers of the clusters as landmarks which represent normal data distribution. For a test data sample, the distance to the nearest landmark gives the outlier score. In the experiments using high dimensional data such as images and documents, it was shown that the proposed method based on the landmarks of one-tenth of training data can give the comparable outlier detection performance while reducing the time complexity greatly in the testing stage.

• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.114-121
• /
• 2019

This paper presents a new benchmark system for visual odometry (VO) and monocular depth estimation (MDE). As deep learning has become a key technology in computer vision, many researchers are trying to apply deep learning to VO and MDE. Just a couple of years ago, they were independently studied in a supervised way, but now they are coupled and trained together in an unsupervised way. However, before designing fancy models and losses, we have to customize datasets to use them for training and testing. After training, the model has to be compared with the existing models, which is also a huge burden. The benchmark provides input dataset ready-to-use for VO and MDE research in 'tfrecords' format and output dataset that includes model checkpoints and inference results of the existing models. It also provides various tools for data formatting, training, and evaluation. In the experiments, the exsiting models were evaluated to verify their performances presented in the corresponding papers and we found that the evaluation result is inferior to the presented performances.

• Journal of the Korean Institute of Gas
• /
• v.21 no.5
• /
• pp.89-94
• /
• 2017

The VR(Virtuality Reality) technology provides very close experience to reality by stimulating humans' external recognition with artificial technologies. In order to overcome the limitation of real-environment training, VR is being applied in industry field as a key technology to prevent safety accident and its control procedure training. However, it is difficult to build VR-based training system because 3D modeling and software coding are necessary for materialization of VR environment demands of many development resource. In this research referring to VR based training content implementation, a method to utilizing VRDC(VR-based Dynamic visualization Component) is suggested and by applying it to plant safety training system, it was confirmed its practicality.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.53-60
• /
• 2012

Nearest-neighbor classification predicts the class of an input data with the most frequent class among the near training data of the input data. Even though nearest-neighbor classification doesn't have a training stage, all of the training data are necessary in a predictive stage and the generalization performance depends on the quality of training data. Therefore, as the training data size increase, a nearest-neighbor classification requires the large amount of memory and the large computation time in prediction. In this paper, we propose a prototype selection algorithm that predicts the class of test data with the new set of prototypes which are near-boundary training data. Based on Tomek links and distance metric, the proposed algorithm selects boundary data and decides whether the selected data is added to the set of prototypes by considering classes and distance relationships. In the experiments, the number of prototypes is much smaller than the size of original training data and we takes advantages of storage reduction and fast prediction in a nearest-neighbor classification.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.777-782
• /
• 2020

As the demand for the talented with computing thinking and programming skills increases the importance of coding education is increasing. Although various coding instructor training programs have been implemented nationwide, little research has been conducted analyzing the current status and contents of coding instructor training programs. Therefore, this thesis presents the design, development process and managing strategies of the 'Coding Instructor Training Courses for 4C Talent Development'. The training program consists of introductory courses and practical coding courses. In the introductory courses, learners acquires the basic knowledge required of coding instructors, and then proceeds to the practical courses to systematically learn the pedagogical knowledge and skills required to educate learners from kindergarten through high school. The case study introduced in this paper is expected to provide useful information to the educators planning and managing the coding instructor training program in the future.

• Journal of The Korean Association of Information Education
• /
• v.26 no.1
• /
• pp.55-64
• /
• 2022

Though school leaders play a crucial role in encouraging the creative school environment, culture, and system that might improve the creative competencies of teachers, there has been little attention to the creative leadership of school supervisors. In this paper, we provide the satisfaction and view of elementary and secondary school leaders who were involved in professional development training for enhancing creative leadership. Using one-way ANOVA and qualitative data coding, we analyze the survey responses from the 67 in-service principals and vice-principals during three different training periods as 2020-I, 2020-II, and 2021-I. We present and compare the result on their satisfaction in terms of satisfaction regarding the process, administrative supports, and optional programs among the training. Next, we explore how the school administrators think the creative leadership training and proposal for its improvement. Based on the feedback for these stakeholders, further studies need to steadily enhance the professional development training for creative leadership and practical program contents.