• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.6
• /
• pp.350-361
• /
• 2019

The fourth industrial revolution based on the intelligent revolution has revolutionized the society as a whole, and it has also affected the defense sector. Various aspects of the war have been changing with the development of technology. In particular, various strategies such as research and development of core technology related to defense unmanned system field and infrastructure are being established based on the fourth industrial revolution technology. In this paper, we have conducted a study to select the technology required for maritime unmanned systems, which can be considered as a priority consideration for the future development of the core technology to be secured prior to the development of the weapon system. First, the core technology prioritization model for the marine unmanned system was established, and the technology fields of the unmanned robot were reclassified and integrated in the related literature such as the classification of the defense technology standard. For the empirical analysis, a questionnaire survey was conducted for 12 specialists who are engaged in the planning of weapons systems, and the importance of technical fields that require development in the development of marine unmanned systems was analyzed. As a result, it was possible to identify the key technology areas that should be considered in selecting the key technologies proposed by the military groups, research institutes, and companies. This could contribute to the establishment of the technology roadmap to develop the marine unmanned system from the future point of view.

• The Korean Journal of Applied Statistics
• /
• v.34 no.5
• /
• pp.745-760
• /
• 2021

Prediction of a stock price has been a subject of interest for a long time in financial markets, and thus, many studies have been conducted in various directions. As the efficient market hypothesis introduced in the 1970s acquired supports, it came to be the majority opinion that it was impossible to predict stock prices. However, recent advances in predictive models have led to new attempts to predict the future prices. Here, we summarize past studies on the price prediction by evaluation measures, and predict the direction of stock prices of Samsung Electronics, LG Chem, and NAVER by applying various machine learning models. In addition to widely used technical indicator variables, accounting indicators such as Price Earning Ratio and Price Book-value Ratio and outputs of the hidden Markov Model are used as predictors. From the results of our analysis, we conclude that no models show significantly better accuracy and it is not possible to predict the direction of stock prices with models used. Considering that the models with extra predictors show relatively high test accuracy, we may expect the possibility of a meaningful improvement in prediction accuracy if proper variables that reflect the opinions and sentiments of investors would be utilized.

• The Journal of Bigdata
• /
• v.6 no.1
• /
• pp.63-81
• /
• 2021

The performance of natural language processing is rapidly improving due to the recent development and application of machine learning and deep learning technologies, and as a result, the field of application is expanding. In particular, as the demand for analysis on unstructured text data increases, interest in NLP(Natural Language Processing) is also increasing. However, due to the complexity and difficulty of the natural language preprocessing process and machine learning and deep learning theories, there are still high barriers to the use of natural language processing. In this paper, for an overall understanding of NLP, by examining the main fields of NLP that are currently being actively researched and the current state of major technologies centered on machine learning and deep learning, We want to provide a foundation to understand and utilize NLP more easily. Therefore, we investigated the change of NLP in AI(artificial intelligence) through the changes of the taxonomy of AI technology. The main areas of NLP which consists of language model, text classification, text generation, document summarization, question answering and machine translation were explained with state of the art deep learning models. In addition, major deep learning models utilized in NLP were explained, and data sets and evaluation measures for performance evaluation were summarized. We hope researchers who want to utilize NLP for various purposes in their field be able to understand the overall technical status and the main technologies of NLP through this paper.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.651-658
• /
• 2021

Because bolts with anti-loosening coatings are used mainly for joining safety-related components in automobiles, accurate automatic screening of these coatings is essential to detect defects efficiently. The performance of the convolutional neural network (CNN) used in a previous study [Identification of bolt coating defects using CNN and Grad-CAM] increased with increasing number of data for the analysis of image patterns and characteristics. On the other hand, obtaining the necessary amount of data for coated bolts is difficult, making training time-consuming. In this paper, resorting to the same VGG16 model as in a previous study, transfer learning was applied to decrease the training time and achieve the same or better accuracy with fewer data. The classifier was trained, considering the number of training data for this study and its similarity with ImageNet data. In conjunction with the fully connected layer, the highest accuracy was achieved (95%). To enhance the performance further, the last convolution layer and the classifier were fine-tuned, which resulted in a 2% increase in accuracy (97%). This shows that the learning time can be reduced by transfer learning and fine-tuning while maintaining a high screening accuracy.

• The Journal of Society for e-Business Studies
• /
• v.26 no.1
• /
• pp.93-106
• /
• 2021

Many recent studies on deep learning-based road crack detection have shown significantly more improved performances than previous works using algorithm-based conventional approaches. However, many deep learning-based studies are still focused on classifying the types of cracks. The classification of crack types is highly anticipated in that it can improve the crack detection process, which is currently relying on manual intervention. However, it is essential to calculate the severity of the cracks as well as identifying the type of cracks in actual pavement maintenance planning, but studies related to road crack detection have not progressed enough to automated calculation of the severity of cracks. In order to calculate the severity of the crack, the type of crack and the area of the crack in the image must be identified together. This study deals with a method of using Mobilenet-SSD that is deep learning-based object detection techniques to effectively automate the simultaneous detection of crack types and crack areas. To improve the accuracy of object-detection for road cracks, several experiments were conducted to combine the U-Net for automatic segmentation of input image and object-detection model, and the results were summarized. As a result, image masking with U-Net is able to maximize object-detection performance with 0.9315 mAP value. While referring the results of this study, it is expected that the automation of the crack detection functionality on pave management system can be further enhanced.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.12
• /
• pp.845-856
• /
• 2020

This study proposes a bridge evaluation system for indivisible permit vehicles such as hydraulic cranes. The permit loads for the bridge evaluation are divided into three categories: routine permit loads, special permit 1 loads, and special permit 2 loads. Routine permit and special permit 1 vehicles are allowed to cross a bridge with normal traffic. For these two permits, the standard lane model in the Korean Highway Bridge Design Code was adopted to consider normal traffic in the same lane. Special permit 2 vehicles are assumed to cross a bridge without other traffic. Structural analyses of two prestressed-beam bridges and two steel box girder bridges were conducted for the proposed permit loads. The rating factors of the four bridges for all permit loads were calculated as sufficiently large values for the moment and shear force so that crossing the bridges can be permitted. A reliability assessment of the bridges was performed to identify the reliability levels for the permit vehicles. It was confirmed that the reliability level of the minimum required strength obtained by the load-resistance factors yields the target reliability index of the design code for the permit vehicles.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.1
• /
• pp.43-50
• /
• 2021

A Leg Mating Unit (LMU) is a device utilized during the float-over installation of offshore structures that include hyperelastic pads and mating part. The hyperelastic pads absorb the loads, whereas the mating part works as guidance between topside and supporting structures during the mating sequence of float-over installation. In this study, the design optimization of an LMU for the float-over installation of floating-type offshore structures is conducted to enhance the performance and to satisfy the requirements defined by classification society regulations. The initial dimensions of the LMU are referred to the dimensions of those used in fixed-type float-over installation because only the location and the number of LMUs are known. The two-parameter Mooney-Rivlin model is adopted to describe the hyperelastic pads under given material parameters. Geometric variables, such as the thickness, height, and width of members, as well as configuration variables, such as the angle and number of members, are defined as design variables and are parameterized. A sampling-based design sensitivity analysis based on latin hypercube sampling method is performed to filter the important design variables. The design optimization problem is formulated to minimize the total mass of the LMU under maximum von Mises stress and reaction force constraints.

• Journal of the Korean Society of Radiology
• /
• v.15 no.4
• /
• pp.547-554
• /
• 2021

In order to quickly and accurately diagnose pneumonia on a chest X-ray image, different batch sizes of 4, 8, 16, and 32 were applied to the same Xception deep learning model, and modeling was performed 3 times, respectively. As a result of the performance evaluation of deep learning modeling, in the case of modeling to which batch size 32 was applied, the results of accuracy, loss function value, mean square error, and learning time per epoch showed the best results. And in the accuracy evaluation of the Test Metric, the modeling applied with batch size 8 showed the best results, and the precision evaluation showed excellent results in all batch sizes. In the recall evaluation, modeling applied with batch size 16 showed the best results, and for F1-score, modeling applied with batch size 16 showed the best results. And the AUC score evaluation was the same for all batch sizes. Based on these results, deep learning modeling with batch size 32 showed high accuracy, stable artificial neural network learning, and excellent speed. It is thought that accurate and rapid lesion detection will be possible if a batch size of 32 is applied in an automatic diagnosis study for feature extraction and classification of pneumonia in chest X-ray images using deep learning in the future.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.321-334
• /
• 2021

In this study, in order to analyze the effect of the consumption value of goods on the purchase intention of corporate brand products, the model was tested using SPSS 21.0 and Smart PLS 3.0 by receiving questionnaires from 457 men and women who have purchased goods. After examining the consumption values of goods, such as limited, functional, aesthetic, hedonistic, innovative, and social ego values, and product satisfaction, the effect on purchase intention of corporate brand products was analyzed. All were found to be significant except for innovative value, and product satisfaction had a significant effect on brand attachment and purchase intention of corporate brand products. We have recently redefine goods according to the trend of the times, and put an academic significance on the first classification of goods into four categories: human brand goods, brand goods, tourism goods, and character goods. It was confirmed that there was a partial difference in consumption value and satisfaction according to the type of goods. Through research, it is intended to broaden the understanding of goods and suggest the possibility of effective and useful marketing by suggesting planning and development directions according to the target.

• Korean Journal of Remote Sensing
• /
• v.38 no.3
• /
• pp.283-298
• /
• 2022

As Synthetic Aperture Radar (SAR) image can be acquired regardless of the weather and day or night, it is highly recommended to be used for Automatic Target Recognition (ATR) in the fields of surveillance, reconnaissance, and national security. However, there are some limitations in terms of cost and operation to build various and vast amounts of target images for the SAR-ATR system. Recently, interest in the development of an ATR system based on simulated SAR images using a target model is increasing. Attributed Scattering Center (ASC) matching and template matching mainly used in SAR-ATR are applied to target classification. The method based on ASC matching was developed by World View Vector (WVV) feature reconstruction and Weighted Bipartite Graph Matching (WBGM). The template matching was carried out by calculating the correlation coefficient between two simulated images reconstructed with adjacent points to each other. For the performance analysis of the two proposed methods, the Synthetic and Measured Paired Labeled Experiment (SAMPLE) dataset was used, which has been recently published by the U.S. Defense Advanced Research Projects Agency (DARPA). We conducted experiments under standard operating conditions, partial target occlusion, and random occlusion. The performance of the ASC matching is generally superior to that of the template matching. Under the standard operating condition, the average recognition rate of the ASC matching is 85.1%, and the rate of the template matching is 74.4%. Also, the ASC matching has less performance variation across 10 targets. The ASC matching performed about 10% higher than the template matching according to the amount of target partial occlusion, and even with 60% random occlusion, the recognition rate was 73.4%.