• Journal of Internet Computing and Services
• /
• v.20 no.1
• /
• pp.77-86
• /
• 2019

As the number of single-person households increases, it is not easy to ask for help alone if a single-person household is severely injured in the home. This paper detects abnormal event when members of a single household in the home are seriously injured. It proposes an vision detection algorithm that analyzes and recognizes patterns through videos that are collected based on home CCTV. And proposes audio detection algorithms that analyze and recognize patterns of sound that occur in households based on Smartphones. If only each algorithm is used, shortcomings exist and it is difficult to detect situations such as serious injuries in a wide area. So I propose a fusion method that effectively combines the two algorithms. The performance of the detection algorithm and the precise detection performance of the proposed fusion method were evaluated, respectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1567-1573
• /
• 2020

The development of Internet technology has brought the era of one-man media. An individual produces content on user own and uploads it to related online services, and many users watch the content of online services using devices that allow them to use the Internet. Currently, most users find and watch content they want through search functions provided by existing online services. These features are provided based on information entered by the user who uploaded the content. In an environment where content needs to be retrieved based on these limited word data, user unwanted information is presented to users in the search results. To solve this problem, in this paper, the system actively analyzes the video in the online service, and presents a way to extract and reflect the characteristics held by the video. The research was conducted to extract morphemes based on the story content based on the voice data of a video and analyze them with big data technology.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.41-50
• /
• 2020

This paper deals with the Braille Education System which complements the shortcomings of the existing Braille Learning Products. An application dedicated to the blind is configured to perform full functions through touch gestures and voice guidance for user convenience. Braille kit is produced for educational purposes through Arduino and 3D printing. The system supports the following functions. First, the learning of the most basic braille, such as initial consonants, final consonant, vowels, abbreviations, etc. Second, the ability to check learned braille by solving step quizzes. Third, translation of braille. Through the experiment, the recognition rate of touch gestures and the accuracy of braille expression were confirmed, and in case of translation, the translation was done as intended. The system allows blind people to learn braille efficiently.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.389-396
• /
• 2022

Deep Learning is based on a perceptron, and is currently being used in various fields such as image recognition, voice recognition, object detection, and drug development. Accordingly, a variety of learning algorithms have been proposed, and the number of neurons constituting a neural network varies greatly among researchers. This study analyzed the learning characteristics according to the number of neurons of the currently used SGD, momentum methods, AdaGrad, RMSProp, and Adam methods. To this end, a neural network was constructed with one input layer, three hidden layers, and one output layer. ReLU was applied to the activation function, cross entropy error (CEE) was applied to the loss function, and MNIST was used for the experimental dataset. As a result, it was concluded that the number of neurons 100-300, the algorithm Adam, and the number of learning (iteraction) 200 would be the most efficient in deep learning learning. This study will provide implications for the algorithm to be developed and the reference value of the number of neurons given new learning data in the future.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.469-484
• /
• 2022

Since a large text corpus contains hundred-thousand unique words, text data is one of the typical large-dimensional data. Therefore, various feature selection methods have been proposed for dimension reduction. Feature selection methods can improve the prediction accuracy. In addition, with reduced data size, computational efficiency also can be achieved. The chi-square statistic and the information gain are two of the most popular measures for identifying interesting terms from text data. In this paper, we investigate the theoretical properties of the chi-square statistic and the information gain. We show that the two filtering metrics share theoretical properties such as non-negativity and convexity. However, they are different from each other in the sense that the information gain is prone to select more negative features than the chi-square statistic in imbalanced text data.

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.9-14
• /
• 2021

In the case of the existing remote collaboration, the remote support service-oriented system is not suitable for the use of the field-oriented multi-person remote collaboration system. This paper is a remote collaboration system development for various industrial sites. We develop remote support and work management that faces the various needs of industrial sites, real-time video remote support between workers, and real-time voice work sharing between workers. In addition, The goal of the development aims to increase the usability by strengthening the security function through encryption in the video and to develop a more efficient system. Finally, the development contents are the remote management and the support software development, Android app development for worker, WebRTC-based remote collaboration system construction and development, and prototype development. These products are expected to increase demand and increase sales by installing and operating at industrial sites, and can promote manpower training, understanding trending technologies, and improving capabilities.

• Journal of Practical Engineering Education
• /
• v.14 no.2
• /
• pp.361-366
• /
• 2022

Recently, with the development of virtual and augmented reality technology, metaverse is emerging as a new paradigm that will lead the next-generation internet era, and social and economic activities are spreading around the game, entertainment, music, and content industries. Moreover, as non-face-to-face conversion accelerated after the outbreak of COVID-19, lifestyles and industrial sites are becoming untact and further rapidly becoming a metaverse. In particular, the application of metaverse to the education field is attracting attention because realistic classes using real-time voice conversations using avatars, 3D objects, and 360-degree images can increase immersion and overcome the limitations of distance education. This study examines the concept of metaverse and examines that education using metaverse can be an alternative that can increase the efficiency of education in the non-face-to-face era. In particular, it shows that it is effective in language education and suggests an actual metaverse-based Korea language education program.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.1
• /
• pp.735-745
• /
• 2023

The purpose of this study is to examine how the control of an agent according to a driver's social status affects user experience factors in a multi-user environment of self-driving vehicles. We conducted a user study where participants viewed four scenarios (route changing/parking x accepting/declining a fellow passenger's command) and answered a survey, followed by a post-hoc interview. Results showed that either the routing scenario or accepting a passenger's command scenario had higher usefulness (convenience, effectiveness, efficiency) than their counterparts. Regardless of the car owner's social status, participants rated AI agents more positively when they met their goals effectively. They also stressed that vehicle owners should always be in control of their agents. This study can provide guidelines for designing future autonomous driving scenarios where an agent interacts with a driver, and passengers.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.10
• /
• pp.935-943
• /
• 2017

CNN (Convolution neural network), which is used for image classification and speech recognition among neural networks learning based on positive data, has been continuously developed to have a high performance structure to date. There are many difficulties to utilize in an embedded system with limited resources. Therefore, we use GPU (General-Purpose Computing on Graphics Processing Units), which is used for general-purpose operation of GPU to solve the problem because we use pre-learned weights but there are still limitations. Since CNN performs simple and iterative operations, the computation speed varies greatly depending on the thread allocation and utilization method in the Single Instruction Multiple Thread (SIMT) based GPGPU. To solve this problem, there is a thread that needs to be relaxed when performing Convolution and Pooling operations with threads. The remaining threads have increased the operation speed by using the method used in the following feature maps and kernel calculations.

• Composites Research
• /
• v.36 no.2
• /
• pp.69-79
• /
• 2023

Oil-water separation is a critical process for several industrial applications, including oil production, wastewater treatment, food processing, and environmental area such as marine oil spills. The separation efficiency of oil-water mixtures can be influenced by various factors such as mixture composition, oil and water conditions, and the separation technology used. Over the years, various technologies have been developed to separate water and oil by physical, chemical and biological methods. This paper presents an overview of the various methods and technologies available for oil-water separation, including gravity separation, centrifugal separation, and separation using adsorbents, filters. The strengths and limitations of each method are discussed, along with recent research trends and future prospects. Furthermore, this paper aims to provide direction for future research and industrial application of sustainable and environmentally friendly oil-water separation technologies. In conclusion, we provide a comprehensive overview of recent oil-water separation technologies that will be beneficial to researchers and industrialists in the field of oil-water separation.