• Journal of the Korea Society of Computer and Information
• /
• v.26 no.5
• /
• pp.1-8
• /
• 2021

Coronavirus disease 2019 (COVID-19) has affected the world seriously. Every person is required for wearing a mask properly in a public area to prevent spreading the virus. However, many people are not wearing a mask properly. In this paper, we propose an efficient mask detection system. In our proposed system, we first detect the faces of input images using YOLOv5 and classify them as the one of three scene complexity classes (Simple, Moderate, and Complex) based on the number of detected faces. After that, the image is fed into the Faster-RCNN with the one of three ResNet (ResNet-18, 50, and 101) as backbone network depending on the scene complexity for detecting the face area and identifying whether the person is wearing the mask properly or not. We evaluated our proposed system using public mask detection datasets. The results show that our proposed system outperforms other models.

• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.285-293
• /
• 2021

With the advent of the 4th Industrial Revolution, AI utilization capabilities are being emphasized in various industries, but AI education design and curriculum research as universal education is currently lacking. This study offers a design for universal AI education to further cultivate its use in universities. For the AI basic education design, a questionnaire was conducted for experts three times, and the reliability of the derived design contents was verified by reflecting the results. As a result, the main competencies for cultivating AI literacy were data literacy, AI understanding and utilization, and the main detailed areas derived were data structure understanding and processing, visualization, word cloud, public data utilization, and machine learning concept understanding and utilization. The educational design content derived through this study is expected to increase the value of competency-centered AI universal education in the future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.7
• /
• pp.869-876
• /
• 2021

In the field of music, various AI composition methods using machine learning have recently been attempted. However, most of this research has been centered on Western music, and little research has been done on Korean traditional music. Therefore, in this paper, we will create a data set of Korean traditional music, create a melody using three algorithms based on the data set, and compare the results. Three models were selected based on the similarity between language and music, LSTM, Music Transformer and Self Attention. Using each of the three models, a melody generator was modeled and trained to generate melodies. As a result of user evaluation, the Self Attention method showed higher preference than the other methods. Data set is very important in AI composition. For this, a Korean traditional music data set was created, and AI composition was attempted with various algorithms, and this is expected to be helpful in future research on AI composition for Korean traditional music.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.3
• /
• pp.35-43
• /
• 2023

Emotion recognition while driving is an essential task to prevent accidents. Furthermore, in the era of autonomous driving, automobiles are the subject of mobility, requiring more emotional communication with drivers, and the emotion recognition market is gradually spreading. Accordingly, in this research plan, the driver's emotions are classified into seven categories using psychological and behavioral data, which are relatively easy to collect. The latent vectors extracted through the auto-encoder model were also used as features in this classification model, confirming that this affected performance improvement. Furthermore, it also confirmed that the performance was improved when using the framework presented in this paper compared to when the existing EEG data were included. Finally, 81% of the driver's emotion classification accuracy and 80% of F1-Score were achieved only through psychological, personal information, and behavioral data.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.23-29
• /
• 2022

Virtual humans are implemented with dedicated modeling tools like Unity 3D Engine in virtual space (virtual reality, mixed reality, metaverse, etc.). Various human modeling tools have been introduced to implement virtual human-like appearance, voice, expression, and behavior similar to real people, and virtual humans implemented via these tools can communicate with users to some extent. However, most of the virtual humans so far have stayed unimodal using only text or speech. As AI technologies advance, the outdated machine-centered dialogue system is now changing to a human-centered, natural multi-modal system. By using several pre-trained networks, we implemented an emotion-based multi-modal dialogue system, which generates human-like utterances and displays appropriate facial expressions in real-time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.4
• /
• pp.691-708
• /
• 2022

As the technology of machine learning and deep learning became common, it began to be applied to research on anomaly(abnormal) detection of industrial control systems. In Korea, the HAI dataset was developed and published to activate artificial intelligence research for abnormal detection of industrial control systems, and an AI contest for detecting industrial control system security threats is being conducted. Most of the anomaly detection studies have been to create a learning model with improved performance through the ensemble model method, which is applied either by modifying the existing deep learning algorithm or by applying it together with other algorithms. In this study, a study was conducted to improve the performance of anomaly detection with a post-processing method that detects abnormal data and corrects the labeling results, rather than the learning algorithm and data pre-processing process. Results It was confirmed that the results were improved by about 10% or more compared to the anomaly detection performance of the existing model.

• Fisheries and Aquatic Sciences
• /
• v.25 no.3
• /
• pp.151-157
• /
• 2022

Various approaches have been applied to transform aquaculture from a manual, labour-intensive industry to one dependent on automation technologies in the era of the fourth industrial revolution. Technologies associated with the monitoring of physical condition have successfully been applied in most aquafarm facilities; however, real-time biological monitoring systems that can observe fish condition and behaviour are still required. In this study, we used a video recorder placed on top of a fish tank to observe the swimming patterns of rock bream (Oplegnathus fasciatus), first one fish alone and then a group of five fish. Rock bream in the video samples were successfully identified using the you-only-look-once v3 algorithm, which is based on the Darknet-53 convolutional neural network. In addition to recordings of swimming behaviour under normal conditions, the swimming patterns of fish under abnormal conditions were recorded on adding an anaesthetic or lowering the salinity. The abnormal conditions led to changes in the velocity of movement (3.8 ± 0.6 cm/s) involving an initial rapid increase in speed (up to 16.5 ± 3.0 cm/s, upon 2-phenoxyethanol treatment) before the fish stopped moving, as well as changing from swimming upright to dying lying on their sides. Machine learning was applied to datasets consisting of normal or abnormal behaviour patterns, to evaluate the fish behaviour. The proposed algorithm showed a high accuracy (98.1%) in discriminating normal and abnormal rock bream behaviour. We conclude that artificial intelligence-based detection of abnormal behaviour can be applied to develop an automatic bio-management system for use in the aquaculture industry.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.179-190
• /
• 2022

Data imbalance refers to a phenomenon in which the number of data in one category is too large or too small compared to another category. Due to this, it has been raised as a major factor that deteriorates performance in machine learning that utilizes classification algorithms. In order to solve the data imbalance problem, various ovrsampling methods for amplifying prime number distribution data have been proposed. Among them, SMOTE is the most representative method. In order to maximize the amplification effect of minority distribution data, various methods have emerged that remove noise included in data (SMOTE-IPF) or enhance only border lines (Borderline SMOTE). This paper proposes a method to ultimately improve classification performance by improving the processing method for anomaly data in the traditional SMOTE method that amplifies minority classification data. The proposed method consistently presented relatively high classification performance compared to the existing methods through experiments.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.287-308
• /
• 2022

The explosive growth of cryptocurrency, led by Bitcoin has emerged as a major issue in the financial market recently. As a result, interest in cryptocurrency investment is increasing, but the market opens 24 hours and 365 days a year, price volatility, and exponentially increasing number of cryptocurrencies are provided as risks to cryptocurrency investors. For that reasons, It is raising the need for research to reduct investors' risks by dividing cryptocurrency which is not suitable for recommendation. Unlike the previous studies of maximizing returns by simply predicting the future of cryptocurrency prices or constructing cryptocurrency portfolios by focusing on returns, this paper reflects the tendencies of investors and presents an appropriate recommendation method with interpretation that can reduct investors' risks by selecting suitable Altcoins which are recommended using Apriori algorithm, one of the machine learning techniques, but based on the similarity and association rules of Bitocoin.

• Computers and Concrete
• /
• v.29 no.6
• /
• pp.375-391
• /
• 2022

This paper examined the impact of the cross-sectional structure on the structural results under different loading conditions of reinforced concrete (RC) members' management limited in Carbon Fiber Reinforced Polymers (CFRP). The mechanical properties of CFRC was investigated, then, totally 32 samples were examined. Test parameters included the cross-sectional shape as square, rectangular and circular with two various aspect rates and loading statues. The loading involved concentrated loading, eccentric loading with a ratio of 0.46 to 0.6 and pure bending. The results of the test revealed that the CFRP increased ductility and load during concentrated processing. A cross sectional shape from 23 to 44 percent was increased in load capacity and from 250 to 350 percent increase in axial deformation in rectangular and circular sections respectively, affecting greatly the accomplishment of load capacity and ductility of the concentrated members. Two Artificial Intelligence Models as Extreme Learning Machine (ELM) and Particle Swarm Optimization (PSO) were used to estimating the tensile and flexural strength of specimen. On the basis of the performance from RMSE and RSQR, C-Shape CFRC was greater tensile and flexural strength than any other FRP composite design. Because of the mechanical anchorage into the matrix, C-shaped CFRCC was noted to have greater fiber-matrix interfacial adhesive strength. However, with the increase of the aspect ratio and fiber volume fraction, the compressive strength of CFRCC was reduced. This possibly was due to the fact that during the blending of each fiber, the volume of air input was increased. In addition, by adding silica fumed to composites, the tensile and flexural strength of CFRCC is greatly improved.