• Journal of Multimedia Information System
• /
• 제7권3호
• /
• pp.205-214
• /
• 2020

Crime has become one of the major problems in modern society. Even though visual surveillances through closed-circuit television (CCTV) is extensively used for solving crime, the number of crimes has not decreased. This is because there is insufficient workforce for performing 24-hour surveillance. In addition, CCTV surveillance by humans is not efficient for detecting dangerous situations owing to accuracy issues. In this paper, we propose the autonomous detection of dangerous situations in CCTV scenes using a deep learning model with relational inference. The main feature of the proposed method is that it can simultaneously perform object detection and relational inference to determine the danger of the situations captured by CCTV. This enables us to efficiently classify dangerous situations by inferring the relationship between detected objects (i.e., distance and position). Experimental results demonstrate that the proposed method outperforms existing methods in terms of the accuracy of image classification and the false alarm rate even when object detection accuracy is low.

• 지식경영연구
• /
• 제19권1호
• /
• pp.175-202
• /
• 2018

The aim of this conceptual article is to explore the sharing economy ecosystem concept in innovation policy context with cluster, innovation system, smart specialization and business ecosystem approaches. This study conducts comparative study to understand what has been changed by sharing economy through Uber case in four cities. By analyzing vital constructs in sharing economy ecosystem, we suggest how sharing economy ecosystem works, and presenting core factors in policy framework of sharing economy ecosystem. In addition, we attempt to explain that policy maker should consider the relationship between these factors. The result of this paper shows sharing economy ecosystem has developed with their characteristics and constructs that are different with traditional industry.

• International journal of advanced smart convergence
• /
• 제8권3호
• /
• pp.167-176
• /
• 2019

Building skins are expanding even beyond theirfunctions as a simple boundary between the exterior and interior and into the realm of linguistic functions thanks to the development of media art. LED has been used as material on outer walls following the advancement of building materials, so the outerskins of large buildings are evolving into a messenger of language capable of communication. In big cities, buildings send out video images to enable communication between people and architecture, which plays a huge role in determining the identity of a building beyond simple advertising. Such media fa?ade technologies can be understood based on the concept of outerskin change, which refers to the idea that animals change the colors or textures of their skins to show their various states. In addition, various message delivery functions in human clothes should be included in such a discussion. We need to research on the possibilities of seeing media facades for their information delivery function and expanding them into information delivery between buildings as well as just between buildings and people.

• Smart Structures and Systems
• /
• 제22권6호
• /
• pp.699-709
• /
• 2018

With increasing traffic volumes and rising vehicle traffic, especially in cities, the number of pedestrian bridges has also increased significantly. Like all other structures, pedestrian bridges also suffer damage. In order to increase the safety of pedestrians, it is necessary to identify existing damage and to repair them to ensure the safety of the bridge structures. Owing to the shortcomings of local methods in identifying damage and in order to enhance the reliability of detection and identification of structural faults, signal methods have seen significant development in recent years. In this research, a new methodology, based on cone-shaped kernel distribution with a new damage index, has been used for damage detection in pedestrian truss bridges. To evaluate the proposed method, the numerical models of the Warren Type steel truss and the Arregar steel footbridge were used. Based on the results, the proposed method and damage index identified the damage and determined its location with a high degree of precision. Given the ease of use, the proposed method can be used to identify faults in pedestrian bridges.

• 전력전자학회논문지
• /
• 제27권3호
• /
• pp.206-213
• /
• 2022

This paper presents setting up a laboratory-scale testbed to estimate the aging of power MOSFET devices and integrated power modules by measuring its on-state voltage and current. Based on the aging mechanisms of the component inside the power module (e.g., bond-wire, solder layer, and semiconductor chip), a system to measure the on-state resistance of device-under-test (DUT) is designed and experimented: a full-bridge circuit applies current stress to DUT, and a temperature chamber controls the ambient temperature of DUT during the aging test. The on-state resistance of SiC MOSFET measured by the proposed testbed was increased by 2.5%-3% after 44-hour of the aging test.

• 전자통신동향분석
• /
• 제36권2호
• /
• pp.12-21
• /
• 2021

Intelligent radar sensors are applied in many industries, such as the automobile, aerospace, and defense industries (for security and surveillance), and for traffic monitoring and management as well as environmental and weather monitoring. Furthermore, they are used in smart cities, homes, and buildings, wherein intelligent motion sensing is required in daily life. It is mentioned that it is being used. In addition, ETRI introduces a phased array-based intelligent radar for drone detection and a human name detection radar technology based on which humans can be detected in case of a disaster.

• ETRI Journal
• /
• 제44권2호
• /
• pp.194-207
• /
• 2022

In the last decade, substantial progress has been achieved in intelligent traffic control technologies to overcome consistent difficulties of traffic congestion and its adverse effect on smart cities. Edge computing is one such advanced progress facilitating real-time data transmission among vehicles and roadside units to mitigate congestion. An edge computing-based deep reinforcement learning system is demonstrated in this study that appropriately designs a multiobjective reward function for optimizing different objectives. The system seeks to overcome the challenge of evaluating actions with a simple numerical reward. The selection of reward functions has a significant impact on agents' ability to acquire the ideal behavior for managing multiple traffic signals in a large-scale road network. To ascertain effective reward functions, the agent is trained withusing the proximal policy optimization method in several deep neural network models, including the state-of-the-art transformer network. The system is verified using both hypothetical scenarios and real-world traffic maps. The comprehensive simulation outcomes demonstrate the potency of the suggested reward functions.

• ETRI Journal
• /
• 제44권2호
• /
• pp.244-259
• /
• 2022

Bicycle- or bike-sharing systems (BSSs) have received increasing attention as a secondary transportation mode due to their advantages, for example, accessibility, prevention of air pollution, and health promotion. However, in BSSs, due to bias in bike demands, the bike rebalancing problem should be solved. Various methods have been proposed to solve this problem; however, it is difficult to apply such methods to small cities because bike demand is sparse, and there are many practical issues to solve. Thus, we propose a demand prediction model using multiple classifiers, time grouping, categorization, weather analysis, and station correlation information. In addition, we analyze real-world relocation data by relocation managers and propose a relocation algorithm based on the analytical results to solve the bike rebalancing problem. The proposed system is compared experimentally with the results obtained by the real relocation managers.

• International Journal of Computer Science & Network Security
• /
• 제23권4호
• /
• pp.25-31
• /
• 2023

Fog computing diversifies cloud computing by using edge devices to provide computing, data storage, communication, management, and control services. As it has a decentralised infrastructure that is capable of amalgamating with cloud computing as well as providing real-time data analysis, it is an emerging method of using multidisciplinary domains for a variety of applications; such as the IoT, Big Data, and smart cities. This present study provides an overview of the security and privacy concerns of fog computing. It also examines its fundamentals and architecture as well as the current trends, challenges, and potential methods of overcoming issues in fog computing.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.431-433
• /
• 2021

본 논문에서는 YOLO 인공 지능 플랫폼을 이용하는 이상행동 감시 시스템을 구현하였으며, YOLO 시스템의 one-shot 감지 시스템 사용으로 기존 감시 시스템에 비해 우수한 응답 특성을 갖는다. YOLO 인공 플랫폼은 폭행, 절도, 방화와 같은 이상행동들로 구성된 이미지 세트로 학습되었다. 이상행동 감시 시스템은 서버와 클라이언트로 구성되어 있으며, 상용화될 경우 각종 범죄 문제를 풀기 위해 스마트시티에 적용이 가능하다.