• International journal of advanced smart convergence
• /
• v.12 no.3
• /
• pp.1-12
• /
• 2023

A public address (PA) system installed in a building is a system that delivers alerts, announcements, instructions, etc. in an emergency or disaster situation. As for the products used in PA systems, with the development of information and communication technology, PA products with various functions have been introduced to the market. PA systems recently launched in the market may be connected through a single network to enable efficient management and operation, or use voice recognition technology to deliver quick information in case of an emergency. In addition, a system capable of locating a user inside a building using a location-based service and guiding or responding to a safe area in the event of an emergency is being launched on the market. However, the new PA systems currently on the market add some functions to the existing PA system configuration to make system operation more convenient, but they do not change the complex PA system configuration to reduce facility costs, maintenance, and management costs. In this paper, we propose a novel PA system configuration for buildings using audio networks and control hierarchy over peer-to-peer (Anchor) technology based on audio over IP (AoIP), which simplifies the complex PA system configuration and enables convenient operation and management. As a result of the study, through the emergency signal processing algorithm, fire broadcasting was made possible according to the detection of the existence of a fire signal in the Anchor system. In addition, the control device of the PA system was replaced with software to reduce the equipment installation cost, and the PA system configuration was simplified. In the future, it is expected that the PA system using Anchor technology will become the standard for PA facilities.

• Journal of the Korea Society for Simulation
• /
• v.19 no.3
• /
• pp.91-98
• /
• 2010

This paper is on the research result for analyzing the performance of GLORY(GLobal Resource management sYstem) used for Web Server Cluster system, which was developed at ETRI(Electronic and Telecommunication Research Institute). The paper includes the definition of Web Server Cluster System, the characteristics of the system, user oriented system performance, current performance enhancement methods, computer simulation model for GLORY and its experimental results for the performance of GLORY. GLORY is composed of 2048~1,000,000 units of PCs, and is used for Internet servers. From the results of the simulation experiments, we notice that GLORY has enough capacity to fully serve the appropriate level of Internet services. Also, the results show that Web server service time is longer than that for network transmission time but requires more DNS than expected, and that 100Mbps LAN is good enough for directly connecting Internet to the Web servers while not affecting the total system performance.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.4
• /
• pp.1-6
• /
• 2023

To enhance immersive experiences for metaverse environements, background music is often used. However, the background music is mostly pre-matched and repeated which might occur a distractive experience to users as it does not align well with rapidly changing user-interactive contents. Thus, we implemented a system to provide a more immersive metaverse conversation experience by 1) developing a regression neural network that extracts emotions from an utterance using KEMDy20, the Korean multimodal emotion dataset 2) selecting music corresponding to the extracted emotions from an utterance by the DEAM dataset where music is tagged with arousal-valence levels 3) combining it with a virtual space where users can have a real-time conversation with avatars.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.2
• /
• pp.201-210
• /
• 2023

Deep learning-based face verificattion model show high performance and are used in many fields, but there is a possibility the user's face image may be leaked in the process of inputting the face image to the model. Althoughde-identification technology exists as a method for minimizing the exposure of face features, there is a problemin that verification performance decreases when the existing technology is applied. In this paper, after combining the face features of other person, a de-identified face image is created through StyleGAN. In addition, we propose a method of optimizingthe combining ratio of features according to the face verification model using HopSkipJumpAttack. We visualize the images generated by the proposed method to check the de-identification performance, and evaluate the ability to maintain the performance of the face verification model through experiments. That is, face verification can be performed using the de-identified image generated through the proposed method, and leakage of face personal information can be prevented.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.4
• /
• pp.1147-1161
• /
• 2023

The Internet of Things (IoT) requires a new processing model that will allow scalability in cloud computing while reducing time delay caused by data transmission within a network. Such a model can be achieved by using resources that are closer to the user, i.e., by relying on edge computing (EC). The amount of IoT data also grows with an increase in the number of IoT devices. However, building such a flexible model within a heterogeneous environment is difficult in terms of resources. Moreover, the increasing demand for IoT services necessitates shortening time delay and response time by achieving effective load balancing. IoT devices are expected to generate huge amounts of data within a short amount of time. They will be dynamically deployed, and IoT services will be provided to EC devices or cloud servers to minimize resource costs while meeting the latency and quality of service (QoS) constraints of IoT applications when IoT devices are at the endpoint. EC is an emerging solution to the data processing problem in IoT. In this study, we improve the load balancing process and distribute resources fairly to tasks, which, in turn, will improve QoS in cloud and reduce processing time, and consequently, response time.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.10
• /
• pp.233-240
• /
• 2009

In this paper we made the serious game learning multiplication table to be evolved. The serious game is to induce the interest of the learner. This program has an interaction form which reflects the intention of the user and using this program a learner to learn the multiplication table as teacher's location are training a program that are seen as the abata and came to be that learner is smart. A study ability to be evolved used an back propagation neural networks. But we improved a study speed using divided weight table concept. The engine is studied perfectly in 60~80 times training. The learning rate increase to various non-monotonic functional form not to do a mechanical rise. And the learning rate is similar with the study ability of the human.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.45-50
• /
• 2023

As the ownership of pets has steadily increased in recent years, the need for an effective pet management system has grown. In this study, we propose a pet management system with a deep learning-based emotion recognition SNS. The system detects emotions through pet facial expressions using a convolutional neural network (CNN) and shares them with a user community through SNS. Through SNS, pet owners can connect with other users, share their experiences, and receive support and advice for pet management. Additionally, the system provides comprehensive pet management, including tracking pet health and vaccination and reservation reminders. Furthermore, we added a function to manage and share pet walking records so that pet owners can share their walking experiences with other users. This study demonstrates the potential of utilizing AI technology to improve pet management systems and enhance the well-being of pets and their owners.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.6 no.12
• /
• pp.533-540
• /
• 2016

Information Visualization is an information technology to visually represent the data and information by utilizing various tools. Visual information can increase the user's intuitiveness and it can be an effective way that can observe and comprehend the specific information in a short time. Information Visualization, as a research method on Humanities, in the high interest of researchers of Humanities and Area studies, is recently used for collecting base data and analyzing information. As an way to use Information Visualization, various forms of Information Visualization like Electronic Cultural Atlas, Network and Multimedia are actively discussed and studied today. In this paper, Information Visualization will be discussed by utilizing D3, a web-based technology for dynamic visualization. In detail, relationship between the Abbasids of Islam can be visualized by using nodes and connection lines and, this relationship between objects will be analyzed efficiently. Researchers related in this field can use this analyzed information for basis data in their researches.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1413-1432
• /
• 2023

Recommender systems aim to recommend items to users by taking into account their probable interests. This study focuses on creating a model that utilizes multiple sources of information about users and items by employing a multimodality approach. The study addresses the task of how to gather information from different sources (modalities) and transform them into a uniform format, resulting in a multi-modal feature description for users and items. This work also aims to transform and represent the features extracted from different modalities so that the information is in a compatible format for integration and contains important, useful information for the prediction model. To achieve this goal, we propose a novel multi-modal recommendation model, which involves extracting latent features of users and items from a utility matrix using matrix factorization techniques. Various transformation techniques are utilized to extract features from other sources of information such as user reviews, item descriptions, and item categories. We also proposed the use of Principal Component Analysis (PCA) and Feature Selection techniques to reduce the data dimension and extract important features as well as remove noisy features to increase the accuracy of the model. We conducted several different experimental models based on different subsets of modalities on the MovieLens and Amazon sub-category datasets. According to the experimental results, the proposed model significantly enhances the accuracy of recommendations when compared to SVD, which is acknowledged as one of the most effective models for recommender systems. Specifically, the proposed model reduces the RMSE by a range of 4.8% to 21.43% and increases the Precision by a range of 2.07% to 26.49% for the Amazon datasets. Similarly, for the MovieLens dataset, the proposed model reduces the RMSE by 45.61% and increases the Precision by 14.06%. Additionally, the experimental results on both datasets demonstrate that combining information from multiple modalities in the proposed model leads to superior outcomes compared to relying on a single type of information.