• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.3
• /
• pp.23-29
• /
• 2019

Conventional crack detection methods have a problem of consuming a lot of labor, time and cost. To solve these problems, an automatic detection system is needed to detect cracks in images obtained by using vehicles or UAVs(unmanned aerial vehicles). In this paper, we have studied road crack detection with unmanned aerial photographs. Aerial images are generated through preprocessing and labeling to generate morphological information data sets of cracks. The generated data set was applied to the mask R-CNN model to obtain a new model in which various crack information was learned. Experimental results show that the cracks in the proposed aerial image were detected with an accuracy of 73.5% and some of them were predicted in a certain type of crack region.

• Electronics and Telecommunications Trends
• /
• v.34 no.4
• /
• pp.15-22
• /
• 2019

Recent technological advances in three-dimensional (3D) sensing devices and machine learning such as deep leaning has enabled data-driven 3D applications. Research on artificial intelligence has developed for the past few years and 3D deep learning has been introduced. This is the result of the availability of high-quality big data, increases in computing power, and development of new algorithms; before the introduction of 3D deep leaning, the main targets for deep learning were one-dimensional (1D) audio files and two-dimensional (2D) images. The research field of deep leaning has extended from discriminative models such as classification/segmentation/reconstruction models to generative models such as those including style transfer and generation of non-existing data. Unlike 2D learning, it is not easy to acquire 3D learning data. Although low-cost 3D data acquisition sensors have become increasingly popular owing to advances in 3D vision technology, the generation/acquisition of 3D data is still very difficult. Even if 3D data can be acquired, post-processing remains a significant problem. Moreover, it is not easy to directly apply existing network models such as convolution networks owing to the various ways in which 3D data is represented. In this paper, we summarize technological trends in AI-based 3D content generation.

• Journal of Korea Technology Innovation Society
• /
• v.22 no.2
• /
• pp.265-281
• /
• 2019

With the advent of the Internet-based economy, the dramatic changes in consumption patterns have been witnessed during the last decades. The seminal change has led by Data Technology, the integrated platform of mobile, online, offline and artificial intelligence, which remained unchallenged. In this paper, I use data analysis tool (TexTom) in order to articulate the definitfite notion of data technology from Internet sources. The data source is collected for last three years (November 2015 ~ November 2018) from Google and Naver. And I have derived several key keywords related to 'Data Technology'. As a result, it was found that the key keyword technologies of Big Data, O2O (Offline-to-Online), AI, IoT (Internet of things), and cloud computing are related to Data Technology. The results of this study can be used as useful information that can be referred to when the Data Technology age comes.

• Clinical and Experimental Pediatrics
• /
• v.62 no.5
• /
• pp.187-192
• /
• 2019

Purpose: The importance of the neurodevelopmental outcomes of very-low-birth-weight (VLBW) infants has been emphasized as their mortality rate has markedly improved. This study aimed to assess the validity of the Korean Developmental Screening Test (K-DST), a developmental screening tool approved by the Korean Society of Pediatrics, for the timely diagnosis of neurodevelopmental delay in VLBW infants. Methods: Subjects included VLBW infants enrolled in the Korean Neonatal Network database between January 2012 and December 2014. The collected data were analyzed for sensitivity, specificity, positive predictive value, and negative predictive value (NPV) in the K-DST compared to those in the Bayley Scales of Infant Development-II for VLBW infants. Results: A total of 173 patients were enrolled. Their mean gestational age and mean birth weight were $27.5{\pm}2.8weeks$ and $980.5{\pm}272.1g$, respectively. The frequency of failed psychomotor developmental index (PDI) <85 was similar to that in at least one domain of K-DST <1 standard deviation. Failure in more than one K-DST domain compared with a mental developmental index (MDI) <85 showed a sensitivity and NPV of 73.2% and 75.0%, respectively. Failure in more than one K-DST domain compared with PDI <85 showed a sensitivity and NPV of 60.3% and 71.6%, respectively. Each K-DST domain had a stronger correlation with predicting a failing MDI <85 than a failing PDI <85 (P<0.05). Conclusion: K-DST could be a useful screening tool for predicting mental developmental delay in VLBW infants and referring them for neurodevelopmental assessments.

• Convergence Security Journal
• /
• v.19 no.1
• /
• pp.145-152
• /
• 2019

In the era of the 3rd Offset Strategy led by the 4th Industrial Revolution, the use of robots with AI and autonomous abilities is becoming more active in military field. The 3rd Offset Strategy is based on the technology of the 4th Industrial Revolution, and S. Korea is heavily dependent on US military technology and is directly and indirectly influenced by the military revolutionary strategy and the alliance relationship. There are many issues that need to be addressed beyond technical maturity for both strategies to be successfully applied in the military. However, there are few discussions about these limitations in many studies and media reports in comparison with the advantages of the techniques. This research describes robot ethics & technology unbalance, problems of autonomous functions, display fatigue induced by VR/AR/MR, cyber/network security to be solved for successful strategies, also the solutions are addressed.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.120-126
• /
• 2019

Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE) has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure (AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificial intelligence (AI) network that can effectively perform pattern recognition of time series data such as the power consumption, and analyze performance of the prediction based on real household power usage data. In the performance analysis, performance comparison between the proposed GRU-based learning model and the conventional learning model of Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absolute error (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE) are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposed GRU-based learning model is greatly improved.

• Journal of The Korean Association of Information Education
• /
• v.25 no.5
• /
• pp.813-824
• /
• 2021

To improve educational performance by analyzing the learning trends of adult learners of Open High Schools, various algorithm models using artificial intelligence were designed and performance was evaluated by applying them to real data. We analyzed Log data of 115 adult learners in the cyber education system of Open High Schools. Most adult learners of Open High Schools learned more than recommended learning time, but at the end of the semester, the actual learning time was significantly reduced compared to the recommended learning time. In the second half of learning, the participation rate of VODs, formation assessments, and learning activities also decreased. Therefore, in order to improve educational performance, learning time should be supported to continue in the second half. In the latter half, we developed an artificial intelligence algorithm models using Tensorflow to predict learning time by data they started taking the course. As a result, when using CNN(Convolutional Neural Network) model to predict single or multiple outputs, the mean-absolute-error is lowest compared to other models.

• Journal of Internet of Things and Convergence
• /
• v.7 no.2
• /
• pp.41-45
• /
• 2021

This paper proposes a video content delivery scheme for vehicles. Today, we spend a lot of time commuting to work in vehicles such as trains and cars. In addition, the number of users who enjoy video content such as YouTube and Netflix in order to appease the boredom in the vehicle is increasing rapidly. Video content requires a larger amount of data usage than text-based content. Hence, the user's mobile communication data usage increases rapidly along with the cost. The proposed video content delivery scheme downloads a lot of video content in advance when the vehicle is in a free Wi-Fi area. In this way, it is possible to play video content in a vehicle at a low cost. It is expected that the proposed scheme can be applied to the Internet of Things(IoT) for moving objects.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.2
• /
• pp.85-92
• /
• 2021

Ambient Air Vaporizer (AAV) is an essential facility in the process of generating natural gas that uses air in the atmosphere as a medium for heat exchange to vaporize liquid natural gas into gas-state gas. AAV is more economical and eco-friendly in that it uses less energy compared to the previously used Submerged vaporizer (SMV) and Open-rack vaporizer (ORV). However, AAV is not often applied to actual processes because it is heavily affected by external environments such as atmospheric temperature and humidity. With insufficient operational experience and facility operations that rely on the intuition of the operator, the actual operation of AAV is very inefficient. To address these challenges, this paper proposes an artificial intelligence-based model that can intelligent AAV operations based on operational big data. The proposed artificial intelligence model is used deep neural networks, and the superiority of the artificial intelligence model is verified through multiple regression analysis and comparison. In this paper, the proposed model simulates based on data collected from real-world processes and compared to existing data, showing a 48.8% decrease in power usage compared to previous data. The techniques proposed in this paper can be used to improve the energy efficiency of the current natural gas generation process, and can be applied to other processes in the future.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.1-6
• /
• 2021

Most of the match-3 puzzle games supports automatic play using the MCTS algorithm. However, implementing reinforcement learning agents is not an easy job because it requires both the knowledge of machine learning and the way of complex interactions within the development environment. This study proposes a method in which we can easily design reinforcement learning agents and implement game play agents by applying PPO(Proximal Policy Optimization) algorithms. And we could identify the performance was increased about 44% than the conventional method. The tools we used are the Unity 3D game engine and Unity ML SDK. The experimental result shows that agents became to learn game rules and make better strategic decisions as experiments go on. On average, the puzzle gameplay agents implemented in this study played puzzle games better than normal people. It is expected that the designed agent could be used to speed up the game level design process.