• Journal of Ocean Engineering and Technology
• /
• v.35 no.1
• /
• pp.91-97
• /
• 2021

Marine accidents caused by ships have brought about economic and social losses as well as human casualties. Most of these accidents are caused by small and medium-sized ships and are due to their poor conditions and insufficient equipment compared with larger vessels. Measures are quickly needed to improve the conditions. This paper discusses a video-integrated collision prediction and fall detection system to support the safe navigation of small- and medium-sized ships. The system predicts the collision of ships and detects falls by crew members using the CCTV, displays the analyzed integrated information using automatic identification system (AIS) messages, and provides alerts for the risks identified. The design consists of an object recognition algorithm, interface module, integrated display module, collision prediction and fall detection module, and an alarm management module. For the basic research, we implemented a deep learning algorithm to recognize the ship and crew from images, and an interface module to manage messages from AIS. To verify the implemented algorithm, we conducted tests using 120 images. Object recognition performance is calculated as mAP by comparing the pre-defined object with the object recognized through the algorithms. As results, the object recognition performance of the ship and the crew were approximately 50.44 mAP and 46.76 mAP each. The interface module showed that messages from the installed AIS were accurately converted according to the international standard. Therefore, we implemented an object recognition algorithm and interface module in the designed collision prediction and fall detection system and validated their usability with testing.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.994-1003
• /
• 2020

In this paper, we investigated the current issues on the KPAS(Korean Public Alert Service) by News analysis. News articles, from May 15, 2005 to April 30, 2020, were collected with the key word of 'KPAS' through the News Big-Data System provided by the Korea Press Foundation. The results of the content analysis are as follows. First, the issues on alert presentation were categorized by alarm sound, message content, alert level, transmission frequency, delay, reception range, time of alert, and language. Issues on inability to receive KPAS messages were categorized into authority, mobile, sending standard, mobile communication infra, etc. For the last two to three years, news on the inability issues had decreased, while news on the presentation issues had increased. This tells us that the public demand for improvement in the KPAS lies in the presentation issues. The demand for societal resolutions to the presentation issues especially on message content, transmission frequency, and reception range has soared.

• Journal of Korea Water Resources Association
• /
• v.54 no.6
• /
• pp.381-393
• /
• 2021

As part of the National Disaster Management Research Institute's Official Development Assistance (ODA) projects for transferring new technologies in the field of disaster-safety management, a flood forecasting and warning system was established in 2019 targeting the Borikhan in the Namxan River Basin in Bolikhamxai Province, Laos. In the target area, which is an ungauged small and medium river basin, observation stations for real-time monitoring of rainfall and runoff and alarm stations were installed, and a software that performs real-time data management and flood forecasting and warning functions was also developed. In order to establish a flood warning standard and develop a nomograph for flood prediction, hydraulic and hydrological analysis was performed based on the 30-year annual maximum daily rainfall data and river morphology survey results in the target area. This paper introduces the process and methodology used in this study, and presents the results of the system's applicability review based on the data observed and collected in 2020 after system installation.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.5
• /
• pp.53-59
• /
• 2018

Due to the recent developments in computer vision technology, complex actions can be recognized with reasonable accuracy in smart cities. In contrast, violence recognition such as events related to fight and knife, has gained less attention. The capability of visual surveillance can be used for detecting fight in streets or in prison centers. In this paper, we proposed a deep learning-based violence recognition method for surveillance cameras. A convolutional neural network (CNN) model is trained and fine-tuned on available benchmark datasets of fights and knives for violence recognition. When an abnormal event is detected, an alarm can be sent to the nearest police station to take immediate action. Moreover, when the probabilities of fight and knife classes are predicted very low, this situation is considered as normal situation. The experimental results of the proposed method outperformed other state-of-the-art CNN models with high margin by achieving maximum 99.21% accuracy.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.2
• /
• pp.45-50
• /
• 2021

With the recent development of IT technology, research and interest in various bio-signal measuring devices are increasing. But studies related to ECG(electrocardiogram), which is one of the most representative bio-signals, particularly arrhythmic signal detection, are incomplete. Since arrhythmia has various causes and has a poor prognosis after onset, preventive treatment through early diagnosis is best. However, the 24-hour Holter electrocardiogram, a tool for diagnosing arrhythmia, has disadvantages in the limitation of use time, difficulty in analyzing motion artifact due to daily life, and the user's real-time alarm function in danger. In this study, an ECG and pulse monitoring device capable of continuous measurement for a long time, a real-time monitoring app, and software for analysis were developed, and the trend of the measured values was confirmed. In future studies, research on derivation of quantitative results of ECG signal measurement analysis is required, and further research on the development of an arrhythmic signal detection algorithm based on this is required.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.6
• /
• pp.137-143
• /
• 2021

In the case of residential facilities, general fire scenarios cannot be applied. Becauseit is difficult to quantify due to the types of combustibles and various fire loads. Existing research conducting surveys of combustibles, but research on fire characteristics is insufficient. Therefore, in this study, an Excel macro that can be quantified by experimenting with the HRR experiments of sofa, drawer, mattress, chair, desk and TV, which are typical combustibles. As a result of experimenting 6 loading combustibles in domestic residential facilities by using a furniture calorimeter, values of 2,391.26kW appeared from the sofa, 1,891.80kW from the drawer, 1,778.95kW from the mattress, 1,104kW from the chair, 291kW from the desk, and 135.09kW from the TV. Also, by applying the α value of the fire growth rate by classifying fire-growing speeds at NFPA 72 (National Fire Alarm Code 2007, Annex B), the mattress can be defined as Very Fast, the sofa and drawer Fast, the TV Slow, the desk Slow, and the chair Medium.

• Journal of the Korean Institute of Gas
• /
• v.25 no.6
• /
• pp.92-97
• /
• 2021

In accordance with the revision of the Enforcement Regulations of the High-Pressure Gas Safety Management Act in February 2021, from August 27, 2021, the operation status of safety devices such as gas leak detection and alarm devices, emergency shut-off devices and flame detectors installed at hydrogen vehicle charging stations can be monitored in real time. It is transmitted and operated by the computer system managed by Korea Gas Safety Corporations. We intend to share the results of statistical analysis of abnormal signals that have occurred along with the results of the monitoring system construction so that they can be used for the safety management of hydrogen refueling stations, and to seek future safety management directions.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.513-518
• /
• 2022

In the past two years, all sectors of society have been underdeveloped and suffered a significant damage due to COVID-19. In particular, people who were working in many areas of research were unable to do anything due to restrictions on meetings with people and social distancing. The purpose of this study is to develop a social distancing device that allows users to maintain the distance in accordance with the government guidelines through an analysis of people's social distancing methods in the COVID-19 society through UX Design. In order to develop the social distancing device, the distance is expressed by changing the color of LED based on research related to UI design such as ultrasonic distance sensor, battery, charging method, distance display method, etc. The outer form of the social distancing device is designed using 3D, the device is developed by installing ultrasonic distance sensor, neo-pixel module and Arduino after printing a prototype using a 3D printer, and this device is tested to develop a final product that helps the social distancing practice amid COVID-19.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.23-31
• /
• 2022

This study analyzed the actual frozen container operation data of Starcool provided by H Shipping. Through interviews with H's field experts, only Critical and Fatal Alarms among the four failure alarms were defined as failures, and it was confirmed that using all variables due to the nature of frozen containers resulted in cost inefficiency. Therefore, this study proposes a method for detecting failure of frozen containers through characteristic importance and PCA techniques. To improve the performance of the model, we select variables based on feature importance through tree series models such as XGBoost and LGBoost, and use PCA to reduce the dimension of the entire variables for each model. The boosting-based XGBoost and LGBoost techniques showed that the results of the model proposed in this study improved the reproduction rate by 0.36 and 0.39 respectively compared to the results of supervised learning using all 62 variables.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1323-1328
• /
• 2021

In this paper, we present a medical electric scooter sharing platform for the transportation vulnerable who are experiencing difficulties and inconveniences in moving. The proposed medical electric scooter sharing platform for the transportation vulnerable includes basic mobile rental, return, and functions that incorporate the IOT technology of the currently operating personal mobility sharing platform. The safety function has been strengthened. The medical electric scooter sharing platform for the transportation vulnerable stores driving data on the server in real time through GPS, and strengthens the alarm and call function in advance of an accident to enable rapid SOS processing. By making the quick contact and responding to the situation, people with disabilities can drive safely and comfortably.