• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.6
• /
• pp.175-181
• /
• 2022

The condition of infrastructure deteriorates as the service life increases. Since most infrastructure in South Korea were intensively built during the period of economic growth, the proportion of outdated infrastructure is rapidly increasing now. Aging of such infrastructure can lead to safety accidents and even human casualties. To prevent these issues in advance, periodic and accurate inspection is essential. For this reason, the need for research to detect various types of damage using computer vision and deep learning is increasingly required in the field of remotely controlled or autonomous inspection. To this end, this study proposed a neural network structure that can detect concrete damage by classifying it into three types. In particular, the proposed neural network can detect them more accurately through a hierarchical learning technique. This neural network was trained with 2,026 damage images and tested with 508 damage images. As a result, we completed an algorithm with average mean intersection over union of 67.04% and F1 score of 52.65%. It is expected that the proposed damage detection algorithm could apply to accurate facility condition diagnosis in the near future.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.30 no.1
• /
• pp.74-81
• /
• 2024

In the near future, autonomous ships, ships controlled by shore remote control centers, and ships operated by navigators will coexist and operate the sea together. In the advent of this situation, a method is required to evaluate the safety of the maritime traffic environment. Therefore, in this study, a plan to evaluate the safety of navigation through ship control simulation was proposed in a maritime environment, where ships directly controlled by navigators and autonomous ships coexisted, using autonomous operation technology. Own ship was designed to have autonomous operational functions by learning the MMG model based on the six-DOF motion with the PPO algorithm, an in-depth reinforcement learning technique. The target ship constructed maritime traffic modeling data based on the maritime traffic data of the sea area to be evaluated and designed autonomous operational functions to be implemented in a simulation space. A numerical model was established by collecting date on tide, wave, current, and wind from the maritime meteorological database. A maritime meteorology model was created based on this and designed to reproduce maritime meteorology on the simulator. Finally, the safety evaluation proposed a system that enabled the risk of collision through vessel traffic flow simulation in ship control simulation while maintaining the existing evaluation method.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.27 no.1
• /
• pp.115-127
• /
• 2024

This research aimed to construct models with various structures based on the Transformer module and to perform land cover classification, thereby examining the applicability of the Transformer module. For the classification of land cover, the Unet model, which has a CNN structure, was selected as the base model, and a total of four deep learning models were constructed by combining both the encoder and decoder parts with the Transformer module. During the training process of the deep learning models, the training was repeated 10 times under the same conditions to evaluate the generalization performance. The evaluation of the classification accuracy of the deep learning models showed that the Model D, which utilized the Transformer module in both the encoder and decoder structures, achieved the highest overall accuracy with an average of approximately 89.4% and a Kappa coefficient average of about 73.2%. In terms of training time, models based on CNN were the most efficient. however, the use of Transformer-based models resulted in an average improvement of 0.5% in classification accuracy based on the Kappa coefficient. It is considered necessary to refine the model by considering various variables such as adjusting hyperparameters and image patch sizes during the integration process with CNN models. A common issue identified in all models during the land cover classification process was the difficulty in detecting small-scale objects. To improve this misclassification phenomenon, it is deemed necessary to explore the use of high-resolution input data and integrate multidimensional data that includes terrain and texture information.

• Proceedings of the Korea Contents Association Conference
• /
• 2003.05a
• /
• pp.92-101
• /
• 2003

The accessibility evaluation of ten web-based loaming content in Korea was performed with the following procedure : (1) A primitive metric of the compliance of those contents to the WCAG (Web Content Accessibility Guidelines) 1.0 was obtained using Bobby, a widely used accessibility checker. (2) SGML validation test was carried out. (3) The contents were rendered with various browsers including a text-mode browser. (4) They were manually checked as to whether they satisfy the accessibility criteria proposed by W3C. Most of the tested contents scored low marks in all the test categories partly because they were apparently developed with little attention paid to web standard conformance, browser compatibility, and device-independence. They also put heavy emphasis on audio-visual effects catering only to the best-equipped users and offering no alternate access route for those in restricted environment. As more information and learning materials are delivered through the Internet, these low accessible contents would lead to a deeper information divide. The accessibility needs to be regarded as an important factor in evaluating the quality of loaming content.

• Journal of Industrial Convergence
• /
• v.18 no.1
• /
• pp.79-85
• /
• 2020

Recently, the RPAS(Remote Piloted Aircraft System), by remote control and autonomous navigation, has been increasing in interest and utilization in various industries and public organizations along with delivery drones, fire drones, ambulances, agricultural drones, and others. The problems of the stability of unmanned drones, which can be self-controlled, are also the biggest challenge to be solved along the development of the drone industry. drones should be able to fly in the specified path the autonomous flight control system sets, and perform automatically an accurate landing at the destination. This study proposes a technique to check arrival by landing point images and control landing at the correct point, compensating for errors in location data of the drone sensors and GPS. Receiving from the Google Map API and learning from the destination video, taking images of the landing point with a drone equipped with a NAVIO2 and Raspberry Pi, camera, sending them to the server, adjusting the location of the drone in line with threshold, Drones can automatically land at the landing point.

• Journal of KIISE:Software and Applications
• /
• v.32 no.7
• /
• pp.674-682
• /
• 2005

This paper proposes classification of multi-spectral satellite image based on fusion of fuzzy G-K (Gustafson-Kessel) algorithm and PCM algorithm. The suggested algorithm establishes the initial cluster centers by selecting training data from each category, and then executes the fuzzy G-K algorithm. PCM algorithm perform using classification result of the fuzzy G-K algorithm. The classification categories are allocated to the corresponding category when the results of classification by fuzzy G-K algorithm and PCM algorithm belong to the same category. If the classification result of two algorithms belongs to the different category, the pixels are allocated by Bayesian maximum likelihood algorithm. Bayesian maximum likelihood algorithm uses the data from the interior of the average intracluster distance. The information of the pixels within the average intracluster distance has a positive normal distribution. It improves classification result by giving a positive effect in Bayesian maximum likelihood algorithm. The proposed method is applied to IKONOS and Landsat TM remote sensing satellite image for the test. As a result, the overall accuracy showed a better outcome than individual Fuzzy G-K algorithm and PCM algorithm or the conventional maximum likelihood classification algorithm.

• 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.2
• /
• pp.265-277
• /
• 2021

With the development of technology, the concept of digital literacy is expanding from the focus of function and tool utilization to the extent of communication and participation in social and cultural contexts. In particular, learners' digital literacy capabilities are more important and necessary in remote learning environments such as Corona19. In this study, we studied how to improve digital literacy education through analysis and inspection of digital literacy shown in elementary school textbooks. To this end, we analyze it through compliance consisting of 8 fundamental pieces of knowledge, 20 sub technical skills, and 10 keys of competencies that constitute digital literacy. As a result of the study, the contents of digital literacy education in elementary school textbooks were presented centered on functions and tools, and as they were biased in certain areas, they were found to be lacking in systematicity and connectivity. Therefore, it proposed the composition of the curriculum system of digital literacy through the independence of elementary informatics curriculum, the specific composition of curriculum, and the strengthening of pre-service teachers' practical skills.

• Journal of Practical Engineering Education
• /
• v.14 no.1
• /
• pp.91-97
• /
• 2022

Recently, as per a representative education method to develop core capabilities (such as critical thinking, communication, collaboration, and creativity) problem based learning (PBL) has been widely adopted in universities. Two important features of PBL are 'collaboration between team members' and 'participation based self-directed learning'. These two features should be satisfied in online education, although it is difficult due to the limitation on space and time in the COVID-19 pandemic. This paper presents a new design of PBL class in Metaverse, based on improving the online PBL class operated in the previous semesters in the H university. In the proposed PBL class, students are able to display materials (e.g., image, pdf, video files) in 3D virtual space, that are related to problem solving. The 3D virtual space is called gallery in this paper. The concept of gallery allows for active participation of students. In addition, the gallery can be used as a tool for collaborative meeting or for final presentation. If possible, the new design of PBL class will be applied and its effectiveness will be analyzed.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.27-39
• /
• 2023

Detecting greenhouses from the remote sensing datasets is useful in identifying the illegal agricultural facilities and predicting the agricultural output of the greenhouses. This research proposed a methodology for automatically detecting greenhouses from a given Planetscope satellite imagery acquired in the areas of Gimje City using the deep learning technique through a series of steps. First, multiple training images with a fixed size that contain the greenhouse features were generated from the five training Planetscope satellite imagery. Next, the YOLO(You Only Look Once) model was trained using the generated training images. Finally, the greenhouse features were detected from the input Planetscope satellite image. Statistical results showed that the 76.4% of the greenhouse features were detected from the input Planetscope satellite imagery by using the trained YOLO model. In future research, the high-resolution satellite imagery with a spatial resolution less than 1m should be used to detect more greenhouse features.