• Journal of Korea Multimedia Society
• /
• v.25 no.3
• /
• pp.493-501
• /
• 2022

This paper is a study on context-sensitive spelling error correction and uses the Korean WordNet (KorLex)[1] that defines the relationship between words as a graph to improve the performance of the correction[2] based on the vector information of the word embedded in the correction technique. The Korean WordNet replaced WordNet[3] developed at Princeton University in the United States and was additionally constructed for Korean. In order to learn a semantic network in graph form or to use it for learned vector information, it is necessary to transform it into a vector form by embedding learning. For transformation, we list the nodes (limited number) in a line format like a sentence in a graph in the form of a network before the training input. One of the learning techniques that use this strategy is Deepwalk[4]. DeepWalk is used to learn graphs between words in the Korean WordNet. The graph embedding information is used in concatenation with the word vector information of the learned language model for correction, and the final correction word is determined by the cosine distance value between the vectors. In this paper, In order to test whether the information of graph embedding affects the improvement of the performance of context- sensitive spelling error correction, a confused word pair was constructed and tested from the perspective of Word Sense Disambiguation(WSD). In the experimental results, the average correction performance of all confused word pairs was improved by 2.24% compared to the baseline correction performance.

• Journal of the Korea Society for Simulation
• /
• v.31 no.2
• /
• pp.11-20
• /
• 2022

There is a critical need for AI assistance in guard operations of Army base perimeters, which is exacerbated by changes in the national defense and security environment such as force reduction. In addition, the possibility for human error inherent to perimeter guard operations attests to the need for an innovative revamp of current systems. The purpose of this study is to propose a real-time object detection AI tailored to military CCTV surveillance with three unique characteristics. First, training data suitable for situations in which relatively small objects must be recognized is used due to the characteristics of military CCTV. Second, we utilize a data augmentation algorithm suited for military context applied in the data preparation step. Third, a noise reduction algorithm is applied to account for military-specific situations, such as camouflaged targets and unfavorable weather conditions. The proposed system has been field-tested in a real-world setting, and its performance has been verified.

• Journal of Creative Information Culture
• /
• v.7 no.2
• /
• pp.99-109
• /
• 2021

As the era of the 4th industrial revolution began and the importance of software emerged, education also reflected this. Software education has already been provided in several countries, and Korea also started software education in the regular curriculum in 2019, when the 2015 revised curriculum was applied. This study attempted to present an improvement plan for revitalizing software education based on the feelings and difficulties of teachers who conducted software education for the first time in the practical education curriculum in elementary school. For the study, a survey was conducted on 96 teachers in charge of software education in elementary schools in 2019 with 36 questions related to personal competency, class operation method, textbooks and educational materials, class operation content, and educational environment. And three of them were interviewed. As a result improvements are needed, such as improving educational facilities and environment, revitalizing the development and dissemination of high-quality instructional materials, and expanding support for participatory training for teachers and teacher clubs.

• Journal of Internet of Things and Convergence
• /
• v.7 no.4
• /
• pp.23-28
• /
• 2021

Through the 4^th industrial revolution, it is rapidly developing in various fields such as artificial intelligence, AR/VR, and big data, and software is at the center. In the field of education as well, the importance of integrated education to support the development of technology is being emphasized, and in order to compete in software technology, securing human resources for software development should be prioritize in domestic. However, unlike the hardware-centric society of the past, the role of software technology human resources is very important, and the reality is that they are discharging human resources that are far from the human resources image that companies need. In this paper, present an immersed education model for training AR software professionals, and based on this, propose an evaluation index that can grasp the quality of the program of the immersed AR education model. Through the AR education model, it is expected that the weaknesses and strengths of the model can be identified, and it can contribute to setting the direction for improvement of the education program.

• Journal of Practical Engineering Education
• /
• v.14 no.2
• /
• pp.279-285
• /
• 2022

In this paper, as a liberal arts course for non-majors, we developed a basic practice case for convolutional neural networks, which is essential for designing a basic convolutional neural network course curriculum. The developed practice case focuses on understanding the working principle of the convolutional neural network and uses a spreadsheet to check the entire visualized process. The developed practice case consisted of generating supervised learning method image training data, implementing the input layer, convolution layer (convolutional layer), pooling layer, and output layer sequentially, and testing the performance of the convolutional neural network on new data. By extending the practice cases developed in this paper, the number of images to be recognized can be expanded, or basic practice cases can be made to create a convolutional neural network that increases the compression rate for high-quality images. Therefore, it can be said that the utility of this convolutional neural network basic practice case is high.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.148-154
• /
• 2021

Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.1
• /
• pp.15-21
• /
• 2023

This paper is a study on the development of an artificial intelligence model that recognizes a hole cup on a golf green. A CNN-based object detection algorithm was used to recognize the hole cup on the green. Also, Apple's CreateML was used to create a model of the object detection algorithm. This paper created a JSON file with 120 training images and annotations to meet the needs of CreateML. In addition, for more accurate learning, data amplification algorithm was used for learning data and 288 learning data were used for learning. By changing the Iterations, Batch size, and Grid size required by CreateML, we found parameter values that improve the performance of the model. A prototype app was developed by applying the developed model, and performance was measured on an actual golf course green using the prototype app. As a result of the measurement, it was found that the hole cup was accurately recognized within 10m, which is the typical golfer's putting distance.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.483-490
• /
• 2022

Malaria is a disease caused by a parasite and it is prevalent in all over the world. The usual method used to recognize malaria cells is a thick and thin blood smears examination methods, but this method requires a lot of manual calculation, so the efficiency and accuracy are very low as well as the lack of pathologists in impoverished country has led to high malaria mortality rates. In this paper, a malaria cell image recognition model using transfer learning is proposed, which consists in the feature extractor, the residual structure and the fully connected layers. When the pre-training parameters of the VGG-19 model are imported to the proposed model, the parameters of some convolutional layers model are frozen and the fine-tuning method is used to fit the data for the model. Also we implement another malaria cell recognition model without residual structure to compare with the proposed model. The simulation results shows that the model using the residual structure gets better performance than the other model without residual structure and the proposed model has the best accuracy of 97.33% compared to other recent papers.

• Imaging Science in Dentistry
• /
• v.52 no.3
• /
• pp.239-244
• /
• 2022

Purpose: Despite the proliferation of numerous morphometric and anthropometric methods for sex identification based on linear, angular, and regional measurements of various parts of the body, these methods are subject to error due to the observer's knowledge and expertise. This study aimed to explore the possibility of automated sex determination using convolutional neural networks(CNNs) based on lateral cephalometric radiographs. Materials and Methods: Lateral cephalometric radiographs of 1,476 Iranian subjects (794 women and 682 men) from 18 to 49 years of age were included. Lateral cephalometric radiographs were considered as a network input and output layer including 2 classes(male and female). Eighty percent of the data was used as a training set and the rest as a test set. Hyperparameter tuning of each network was done after preprocessing and data augmentation steps. The predictive performance of different architectures (DenseNet, ResNet, and VGG) was evaluated based on their accuracy in test sets. Results: The CNN based on the DenseNet121 architecture, with an overall accuracy of 90%, had the best predictive power in sex determination. The prediction accuracy of this model was almost equal for men and women. Furthermore, with all architectures, the use of transfer learning improved predictive performance. Conclusion: The results confirmed that a CNN could predict a person's sex with high accuracy. This prediction was independent of human bias because feature extraction was done automatically. However, for more accurate sex determination on a wider scale, further studies with larger sample sizes are desirable.

• Proceedings of the KIPE Conference
• /
• 1998.10a
• /
• pp.961-965
• /
• 1998

The rapid development of small-scale (1-10 MJ) Superconducting Magnetic Energy Storage Systems (SMES) can be explained by real perspective of practical implementation of these devices in electro power nets. However the serious problem of all high mechanically stressed superconducting coils-problem of training and degradation (decreasing) of operating current still exists. Moreover for SMES systems this problems is more dangerous because of pulsed origin of mechanical stresses-one of the major sources of local heat disturbances in superconducting coils. We investigated acoustic emission (AE) phenomenon on model and 0.5 MJ SMES coils taking into account close correlation of AE and local heat disturbances. Two-coils 0.5 MJ SMES system was developed, manufactured and tested at Russian Research Center in the frames of cooperation with Korean Electrical Engineering Company (KEPCO) [1]. The two-coil SMES operates with the stored energy transmitted between coils in the course of a single cycle with 2 seconds energy transfer time. Maximum operating current 1.55 kA corresponds to 0.5 MF in each coil. The Nb-Ti-based conductor was designed and used for SMES manufacturing. It represents transposed cable made of Nb-Ti strands in copper matrix, several cooper strands and several stainless steel strands. The coils are wound onto fiberglass cylindrical bobbins. To make AE event information more useful a real time instrumentation system was used. Two main measured and computer processed AE parameters were considered: the energy of AE events (E) and the accumulated energy of AE events (E ). Influence of current value in 0.5 MJ coils on E and E was studied. The sensors were installed onto the bobbin and the external surface of magnets. Three levels of initial current were examined: 600A, 1000A, 2450 A. An extraordinary strong dependence of the current level on E and E was observed. The specific features of AE from model coils, operated in sinusoidal vibration current changing mode were investigated. Three current frequency modes were examined: 0.012 Hz, 0.03 Hz and 0.12 Hz. In all modes maximum amplitude 1200 A was realized.