• Journal of the Korea Society for Simulation
• /
• v.31 no.1
• /
• pp.29-41
• /
• 2022

As one of the alternatives to solve the problem of unstable food supply and demand imbalance caused by abnormal climate change, the need for plant factories is increasing. Airflow in plant factory is recognized as one of important factor of plant which influence transpiration and heat transfer. On the other hand, Digital Twin (DT) is getting attention as a means of providing various services that are impossible only with the real system by replicating the real system in the virtual world. This study aimed to develop a digital twin model for airflow prediction that can predict airflow in various situations by applying the concept of digital twin to a plant factory in operation. To this end, first, the mathematical formalism of the digital twin model for airflow analysis in plant factories is presented, and based on this, the information necessary for airflow prediction modeling of a plant factory in operation is specified. Then, the shape of the plant factory is implemented in CAD and the DT model is developed by combining the computational fluid dynamics (CFD) components for airflow behavior analysis. Finally, the DT model for high-accuracy airflow prediction is completed through the validation of the model and the machine learning-based calibration process by comparing the simulation analysis result of the DT model with the actual airflow value collected from the plant factory.

• Cartoon and Animation Studies
• /
• s.47
• /
• pp.273-294
• /
• 2017

Korea has suffered numerous casualties due to a lot of accidents caused by safety insufficiency in recent years. Therefore, safety education is more important than ever before, and 'how to educate with what contents' is an important subject. Especially, experience education is effective rather than theoretical education because of the nature of safety education. However, it is not easy to design and develop these safety education programs. There is not much opportunity to access first-aid training, which is a part of safety education, unless it is compulsory to learn through public institutions. As a result, program utilization on safety education in Korea is still insufficient to what it should be. With that taken into account, this study proposed an effective serious game with fun and immersion for medical first-aid education. To do this, we analyzed five medical games through 20 cases of first-aid applications and elicited five factors that enhance the usability of serious games. With an analysis of five medical games, we selected one game to borrow the game rules, and applied the elicited five elements in the forms of level-up structure, iterative learning, compensation outcomes, competition system, and information transfer. The proposed medical education functional games should have 1) a character that plays a role of a patient, 2) a narrative flow that shows the situation, 3) the user should judge the situation and induce first aid. 4) compensation, levels, and simple repetition should be designed, and 5) information should be shared with the others in the given community. The results of this study is believed to contribute to enhance the medical emergency education in Korea.

• Journal of Appropriate Technology
• /
• v.6 no.2
• /
• pp.200-207
• /
• 2020

The purpose of the establishment of Fab Lab is to promote the sustainable development of local communities around the world. To this end, The Fab foundation are preparing a resource-circulating society that maintains a city's self-sufficiency rate of 50% or more by 2054. In developed countries, Fab Lab is not only a manufacturing space for startup support, but an open innovation space for learning and creation. In addition, in emerging countries, Fab Lab is playing a role as a digital production center to create and share appropriate new technologies by reflecting the needs of local communities. India has 70 Fab Labs, the largest emerging country, ahead of Russia's 48. India's Fab Lab is conducting a collaboration project through regular meetings held every six months. The subject of this study, Fab Lab Vigyan Ashram, is defined as a place to transfer the concept of digital lab to alternative schools in rural India. In this study, we looked at a case in which an alternative school for an agricultural community called Vigyan Ashram, the modern version of the Gurukula system, successfully combined with the digital fabrication called Fab Lab to become a new citizen-led making community of the 4th Industrial Revolution. Based on this, we explored the development model of the Indian Fab Lab that fits the local situation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.32-39
• /
• 2018

Raman spectra exhibit differences in intensity depending on the measuring equipment and environmental conditions even for the same material. This restricts the pattern recognition approach of Raman spectroscopy and is an issue that must be solved for the sake of its practical application, so as to enable the reusability of the Raman database and interoperability between Raman devices. To this end, previous studies assumed the existence of a transfer function between the measurement devices to obtain a direct spectral correction. However, this method cannot cope with other conditions that cause various intensity distortions. Therefore, we propose a classification method using linear intensity calibration which can deal with various measurement conditions more flexibly. In order to evaluate the performance of the proposed method, a Raman library containing 14033 chemical substances was used for identification. Ten kinds of chemical Raman spectra measured using three different Raman spectroscopes were used as the experimental data. The experimental results show that the proposed method achieves 100% discrimination performance against the intensity-distorted spectra and shows a high correlation score for the identified material, thus making it a useful tool for the identification of chemical substances.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.165-172
• /
• 2020

In this paper we propose an uses on-device-based edge computing technology and big data analysis methods through the use of on-device-based edge computing technology and analysis of big data, which are distributed computing paradigms that introduce computations and storage devices where necessary to solve problems such as transmission delays that occur when data is transmitted to central centers and processed in current general smart factories. However, even if edge computing-based technology is applied in practice, the increase in devices on the network edge will result in large amounts of data being transferred to the data center, resulting in the network band reaching its limits, which, despite the improvement of network technology, does not guarantee acceptable transfer speeds and response times, which are critical requirements for many applications. It provides the basis for developing into an AI-based facility prediction conservation analysis tool that can apply deep learning suitable for big data in the future by supporting intelligent facility management that can support productivity growth through research that can be applied to the field of facility preservation and smart factory industry with integrated hardware technology that can accommodate these requirements and factory management and control technology.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.343-350
• /
• 2022

It is worth verifying the effectiveness of data integration between data with different features. This study investigated whether the data integration affects the accuracy of deep neural network (DNN), and which integration method shows the best improvement. This study used two different public datasets. One public dataset was taken in an actual farm in India. And another was taken in a laboratory environment in Korea. Leaf images were selected from two different public datasets to have five classes which includes normal and four different types of plant diseases. DNN used pre-trained VGG16 as a feature extractor and multi-layer perceptron as a classifier. Data were integrated into three different ways to be used for the training process. DNN was trained in a supervised manner via the integrated data. The trained DNN was evaluated by using a test dataset taken in an actual farm. DNN shows the best accuracy for the test dataset when DNN was first trained by images taken in the laboratory environment and then trained by images taken in the actual farm. The results show that data integration between plant images taken in a different environment helps improve the performance of deep neural networks. And the results also confirmed that independent use of plant images taken in different environments during the training process is more effective in improving the performance of DNN.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.21-29
• /
• 2023

In this paper, we propose an AI-based system for automatically classifying industry and occupation codes in the population census. The accurate classification of industry and occupation codes is crucial for informing policy decisions, allocating resources, and conducting research. However, this task has traditionally been performed by human coders, which is time-consuming, resource-intensive, and prone to errors. Our system represents a significant improvement over the existing rule-based system used by the statistics agency, which relies on user-entered data for code classification. In this paper, we trained and evaluated several models, and developed an ensemble model that achieved an 86.76% match accuracy in industry and 81.84% in occupation, outperforming the best individual model. Additionally, we propose process improvement work based on the classification probability results of the model. Our proposed method utilizes an ensemble model that combines transfer learning techniques with pre-trained models. In this paper, we demonstrate the potential for AI-based systems to improve the accuracy and efficiency of population census data classification. By automating this process with AI, we can achieve more accurate and consistent results while reducing the workload on agency staff.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.77-89
• /
• 2022

It has been proven through many previous researches that the pretrained language model with a large corpus helps improve performance in various natural language processing tasks. However, there is a limit to building a large-capacity corpus for training in a language environment where resources are scarce. Using the Cross-lingual Post-Training (XPT) method, we analyze the method's efficiency in Korean, which is a low resource language. XPT selectively reuses the English pretrained language model parameters, which is a high resource and uses an adaptation layer to learn the relationship between the two languages. This confirmed that only a small amount of the target language dataset in the relationship extraction shows better performance than the target pretrained language model. In addition, we analyze the characteristics of each model on the Korean language model and the Korean multilingual model disclosed by domestic and foreign researchers and companies.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.8
• /
• pp.5369-5377
• /
• 2014

Korean pagodas were constructed in the shape of a wood pagoda, brick pagoda, stone pagoda, etc. On the other hand, the currently remaining traditional pagodas are those having nonflammable materials, such as brick, stone, etc. Compared to the stone pagoda, there is data regarding brick pagodas, but there is little literature data on how to construct these pagodas. This appears to be because there are relatively few Korean brick pagodas currently remaining, they are locally restricted, the material limit is not overcome, pagoda's historical and regional problems have not been analyzed, and pagoda construction is centered on pagoda construction. Therefore, this study examined the local cultural characteristics on the construction of brick pagodas. As a result, cultural exchange between Korea and China was performed through the silk road and there was a marine route for cultural exchange. Such exchange was shared with the East Asia area as well, which can be found by comparing remains at related areas. Exchange with China can be mentioned as the selective exchange of local powers as well as blind learning. Second, brick pagoda were constructed in Korea because of the good quality soil easily. Uisang's Hwaeomjong was negotiated with the main power not agreeing with Buddhism, which was popularized and the local power. Third, brick pagoda construction was influenced by negotiation related between Balhae and Silla, in which the ethnic influence was locally affected and could be mentioned as being a culturally selective result transferred from China. As a result, brick pagodas can be oriented by forming a unitary state rather than a small country within China's influence range as well as cultural transfer through the silk road.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.6
• /
• pp.1137-1144
• /
• 2022

In this study, an artificial intelligence(AI) was developed to help with facial expression practice in order to express emotions. The developed AI used multimodal inputs consisting of sentences and facial images for deep neural networks (DNNs). The DNNs calculated similarities between the emotions predicted by the sentences and the emotions predicted by facial images. The user practiced facial expressions based on the situation given by sentences, and the AI provided the user with numerical feedback based on the similarity between the emotion predicted by sentence and the emotion predicted by facial expression. ResNet34 structure was trained on FER2013 public data to predict emotions from facial images. To predict emotions in sentences, KoBERT model was trained in transfer learning manner using the conversational speech dataset for emotion classification opened to the public by AIHub. The DNN that predicts emotions from the facial images demonstrated 65% accuracy, which is comparable to human emotional classification ability. The DNN that predicts emotions from the sentences achieved 90% accuracy. The performance of the developed AI was evaluated through experiments with changing facial expressions in which an ordinary person was participated.