• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.129-131
• /
• 2022

본 논문에서는 열차의 운행으로 인한 철도 주변 주거지역에서 다양한 유형의 위험요소를 파악하고 분석하기 위한 시스템 운영방안을 제안한다. 위험 요소를 파악하고, 특정 위치의 필요한 센서를 부착하여 데이터를 수집 및 처리하고 패턴을 분석하여 사용자에게 필요한 정보를 제공함으로써, 철도 주변 주거지역에 어떠한 피해가 있는지 알 수 있고, 그에 적합한 적용방안을 마련하고, 시스템 제어를 위한 애플리케이션과 연동하여 사용자에게 더 나은 편의성을 제공할 수 있다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.289-292
• /
• 2022

본 논문에서는 실제 철도 건널목(교차로)에서 발생하는 소음 및 진동, 차량 및 보행자 사고와 같은 위험 요소로부터 발생하는 위험 상황들을 분류하고, 철도 건널목(교차로) 운행 상황을 구현한 모형 철도 주변에 센서를 부착하여 철도 건널목에서 발생하는 위험 요소들을 아두이노 센서로 감지해 데이터를 수집한다. 또한 수집된 데이터들을 활용하여 사용자의 상황에 맞는 철도데이터 운영시스템을 제안한다.

• Journal of Korean Society of Transportation
• /
• v.33 no.4
• /
• pp.323-336
• /
• 2015

Estimating the risks on the roadway using surrogate safety measures (SSM) has an advantage in that it focuses on the vehicle trajectory directly involved in conflicts. On the other hand, there is a restriction on estimating the risks of continuous segments due to the limited data collected from a location. To overcome the restriction, this study presents the scheme of acquiring the vehicular trajectory using real time kinematics-differential global positioning system (RTK-DGPS) and develops a methodology which contains the considerations of the problems to calculate the SSM such as time-to-collision (TTC), deceleration rate to avoid collision (DRAC) and acceleration noise (AN). By using the methodology, this study shows a result from an experiment executed in a section where the variation of vehicular movement can be observed from several continuous flow roadway sections near Seoul and Gyeonggi Province in Korea. The result illustrated the risks on the roadway by the SSM metrics in certain situations like merging and diverging, stop-and-go, and weaving. This study would be applied to relate the dangers with characteristics of drivers and roadway sections, and prevenst accidents or conflicts by detecting dangerous roadway sections and drivers' behaviors. This study contributes to improving roadway safety and reducing car-accidents.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.

• Journal of Advanced Navigation Technology
• /
• v.21 no.1
• /
• pp.126-131
• /
• 2017

This paper designs a smart safety measurement system with Bluetooth beacon sensor nodes that can provide risk detection and evacuation/countermeasure services. The Bluetooth beacon sensor nodes is easily able to be attached to old building wall or construction or civil structure with potential danger. The proposed smart safety measurement system transmits various sensor data such as acceleration, gyroscope, geomagnetic, pressure, altitude, temperature, humidity at the spot where Bluetooth beacon sensor nodes are installed, and we can use them for risk perception, prediction, and warning services. To verify the effectiveness of the proposed system, we performed filed tests which showed that measured displacement values of precast retaining walls were within the permitted displacement value of 38.5 mm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1259-1265
• /
• 2022

Recently, AI, big data, and IoT technologies are being used in various solutions such as fire detection and gas or dangerous substance detection for safety accident prevention. According to the status of occupational accidents published by the Ministry of Employment and Labor in 2021, the accident rate, the number of injured, and the number of deaths have increased compared to 2020. In this paper, referring to the dataset construction guidelines provided by the National Intelligence Service Agency(NIA), the dataset is directly collected from the field and learned with YOLOv4 to propose a collision risk object detection system through object detection. The accuracy of the dangerous situation rule violation was 88% indoors and 92% outdoors. Through this system, it is thought that it will be possible to analyze safety accidents that occur in industrial sites in advance and use them to intelligent platforms research.

• Journal of the Korean Institute of Gas
• /
• v.27 no.1
• /
• pp.63-69
• /
• 2023

Although Korea is the world's No. 1 semiconductor producing country, most studies are conducted with risk assessment for simple material risks due to the closedness of the site for industrial protection. In terms of industrial safety, a monitoring system such as a gas detector to determine the leakage of hazardous substances has been established, but research on effectively discharging harmful gastritis substances in case of leakage has only recently begun. Semiconductor manufacturing facilities (gas boxes) where a large amount of flammable materials are handled are currently being safety managed by using a gas detector and blocking the air inlet. It is difficult to dilute in a short time in case of leakage of flammable substances. Therefore, in this study, based on various criteria, the size of the duct according to the size of the gas box is determined and the appropriate size of the air inlet is studied to minimize the exhaust performance requirement without exposing hazardous chemicals to the outside in the event of a flammable leak. We want to do an optimal exhaust design.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.1
• /
• pp.11-18
• /
• 2022

The use of multi-spectral cameras is essential for day and night pedestrian detection. In this paper, a color camera and a thermal imaging infrared camera were used to detect pedestrians near a crosswalk for 24 hours at an intersection with a high risk of traffic accidents. For pedestrian detection, the YOLOv5 object detector was used, and the detection performance was improved by using color images and thermal images at the same time. The proposed system showed a high performance of 0.940 mAP in the day/night multi-spectral (color and thermal image) pedestrian dataset obtained from the actual crosswalk site.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.225-231
• /
• 2022

최근 기계번역 분야는 괄목할만한 발전을 보였으나, 번역 결과의 오류가 불완전한 의미의 왜곡으로 이어지면서 사용자로 하여금 불편한 반응을 야기하거나 사회적 파장을 초래하는 경우가 존재한다. 특히나 오역에 의해 변질된 의미로 인한 경제적 손실 및 위법 가능성, 안전에 대한 잘못된 정보 제공의 위험, 종교나 인종 또는 성차별적 발언에 의한 파장은 실생활과 문제가 직결된다. 이러한 문제를 완화하기 위해, 기계번역 품질 예측 분야에서는 치명적 오류 감지(Critical Error Detection, CED)에 대한 연구가 이루어지고 있다. 그러나 한국어에 관련해서는 연구가 존재하지 않으며, 관련 데이터셋 또한 공개된 바가 없다. AI 기술 수준이 높아지면서 다양한 사회, 윤리적 요소들을 고려하는 것은 필수이며, 한국어에서도 왜곡된 번역의 무분별한 증식을 낮출 수 있도록 CED 기술이 반드시 도입되어야 한다. 이에 본 논문에서는 영어-한국어 기계번역 분야에서의 치명적 오류를 감지하는 KoCED(English-Korean Critical Error Detection) 데이터셋을 구축 및 공개하고자 한다. 또한 구축한 KoCED 데이터셋에 대한 면밀한 통계 분석 및 다국어 언어모델을 활용한 데이터셋의 타당성 실험을 수행함으로써 제안하는 데이터셋의 효용성을 면밀하게 검증한다.

• Journal of IKEEE
• /
• v.22 no.1
• /
• pp.168-173
• /
• 2018

To overcome the problems of the existing black box which is exposed to the risk of blind spots in the imaging of a fixed front and rear views of an object, this paper suggests a new intelligent black box that can detect and shoot side views of an object. This paper proposes an algorithm of the intelligent black box with a rotating function in order to compensate for the side blind spot of the vehicle. This intelligent black box with rotating screen adopts the infrared distance sensor to sense an object which approaches to the vehicle and rotates automatically towards the object.