• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1165-1171
• /
• 2022

Human behavior recognition using an accelerometer has been applied to various fields. As smartphones have become used commonly, a method for human behavior recognition using the acceleration sensor built into the smartphone is being studied. In the case of the elderly, falling often leads to serious injuries, and falls are one of the major causes of accidents at construction fields. In this article, we proposed recognition method for human falling direction using built-in acceleration sensor and orientation sensor in the smartphone. In the past, it was a common method to use the magnitude of the acceleration vector to recognize human behavior. These days, deep learning has been actively studied and applied to various areas. In this article, we propose a method for recognizing the direction of human falling by applying the deep learning multilayer technique, which has been widely used recently.

• Journal of Sensor Science and Technology
• /
• v.31 no.2
• /
• pp.96-101
• /
• 2022

As the elderly population gradually increases, the risk of fatal fall accidents among the elderly is increasing. One way to cope with a fall accident is to determine the fall direction before impact using a wearable inertial measurement unit (IMU). In this context, a previous study proposed a method of classifying fall directions using a support vector machine with sensor velocity, acceleration, and tilt angle as input parameters. However, in this method, the IMU signals are processed through several processes, including a Kalman filter and the integration of acceleration, which involves a large amount of computation and error factors. Therefore, this paper proposes a machine learning-based method that classifies the fall direction before impact using IMU raw signals rather than processed data. In this study, we investigated the effects of the following two factors on the classification performance: (1) the usage of processed/raw signals and (2) the selection of machine learning techniques. First, as a result of comparing the processed/raw signals, the difference in sensitivities between the two methods was within 5%, indicating an equivalent level of classification performance. Second, as a result of comparing six machine learning techniques, K-nearest neighbor and naive Bayes exhibited excellent performance with a sensitivity of 86.0% and 84.1%, respectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.252-254
• /
• 2021

According to the Institute for Occupational Safety and Health, the number of industrial injuries in 2019 was 109,242, an increase of 6.8% from 2018. In this situation, the government and companies are discussing the development of core technologies for preventing safety accidents on site based on ICT in the field of construction and construction. In these fields, technologies using computer vision and artificial intelligence have recently been widely used. In this paper, we built training data for safety management of factory workers and trained a model based on YOLOv4. It is believed that this can be used as an initial study to predict the risk situation of workers in factories.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.505-507
• /
• 2021

Image recognition technology is a technology that recognizes an image object by using the generated feature descriptor and generates object feature points and feature descriptors that can compensate for the shape of the object to be recognized based on artificial intelligence technology, environmental changes around the object, and the deterioration of recognition ability by object rotation. The purpose of the present invention is to implement a power management framework required to increase profits and minimize damage to livestock farmers by preventing accidents that may occur due to the improvement of efficiency of the use of livestock house power and overloading of electricity by integrating and managing a power fire management device installed for analyzing a complex environment of power consumption and fire occurrence in a smart safety livestock house, and to develop and disseminate a safe and optimized intelligent smart safety livestock house.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.815-817
• /
• 2022

This paper is about the development of an automatic stair climbing trolley for carrying loads without manpower. The design of tri-wheeled structure and center of mass enable the trolley to move on flat ground and also to ascend stairs by self-balancing. The overall design enables the trolley to avoid collision to walls when the trolley rotates on domestic landings. When the camera recognizes the stair, the sensor measures distance from the trolley to the stair. Then the trolley can move to align itself in the middle of the stair and it starts climbing. It can ascend to a specific floor based on the floor number entered by the user. As a result, the automatic stair climbing trolley is expected to help humans by protecting from accidents of dropping loads and saving their power. It is also expected to use for various purposes such as delivering packages, moving and carrying heavy loads in buildings without elevator.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.12
• /
• pp.8879-8888
• /
• 2015

When the disaster like earthquake in urban area occur, due to the collapse accidents for subway, tunnel space with buildings or underground area, enormous property and human damage are happened. Specially, since it is difficult to identify survived status of humans within collapsed debris and accurately buried locations of the humans, inputs of considerable time and manpower for rescuing them are required. Besides, secondary damage can be occurred by additional collapses. The aim of this study is to propose a stochastic location positioning method that enables to provide aid information by determining locations of mobile devices for buried persons in 2-D plane using wireless communication technologies. This study selected a detection method for buried persons based on Wi-Fi signal, and identified characteristics of signal strengths by distance unit. Using these methods, a stochastic location detection model in 2-D plane was built. It is expected that this technology will be utilized as a core technology that can protects safety and human life of the public by providing data for rescuing quickly buried persons in cases of national disasters for future.

• Journal of Korean Society of Transportation
• /
• v.32 no.4
• /
• pp.293-302
• /
• 2014

Recently, advanced sensors and communication technologies have been widely applied to advanced safety vehicles for reducing traffic accidents and injury severity. To apply the advanced safety vehicle technologies, it is important to quantify safety benefits, which is a fundamental for justifying application. This study proposed a methodology for quantifying the effectiveness of the Advanced Driver Assistant System (ADAS) with the Analytic Hierarchy Process (AHP). When the proposed methodology is applied to 2008-2010 Gyeonggi-province crash data, ADAS would reduce about 10.18% of crashes. In addition, Adaptive Cruise Control, Automatic Emergency Braking System, Lane Departure Warning System and Blind Spot Detection System are expected to reduce about 10.43%, 10.17%, 9.96%, and 10.18%, respectively. The outcomes of this study might support decision making for developing not only vehicular technologies but also relevant safety policies.

• Journal of Korea Water Resources Association
• /
• v.53 no.4
• /
• pp.313-322
• /
• 2020

There is always a risk of water disasters due to sudden storms in mountainous regions in Korea, which is more than 70% of the country's land. In this study, a radar-based risk prediction technique for sudden downpour is applied in the mountainous region and is evaluated for its applicability using Mt. Biseul rain radar. Eight local heavy rain events in mountain regions are selected and the information was calculated such as early detection of cumulonimbus convective cells, automatic detection of convective cells, and risk index of detected convective cells using the three-dimensional radar reflectivity, rainfall intensity, and doppler wind speed. As a result, it was possible to confirm the initial detection timing and location of convective cells that may develop as a localized heavy rain, and the magnitude and location of the risk determined according to whether or not vortices were generated. In particular, it was confirmed that the ground rain gauge network has limitations in detecting heavy rains that develop locally in a narrow area. Besides, it is possible to secure a time of at least 10 minutes to a maximum of 65 minutes until the maximum rainfall intensity occurs at the time of obtaining the risk information. Therefore, it would be useful as information to prevent flash flooding disaster and marooned accidents caused by heavy rain in the mountainous area using this technique.

• Journal of the Korean Institute of Gas
• /
• v.21 no.2
• /
• pp.41-49
• /
• 2017

Recently, as the number of facilities using chemicals is increasing, the amount of handling is rapidly increasing. However, chemical spills are occurring steadily, and if large quantities of chemicals are leaked in time, they are likely to cause major damage. These industrial complexes use information obtained from a number of sensors to detect and monitor leaking areas, and are used in industrial fields by applying existing fixed sensors to robots and drones. Therefore, it is necessary to propose a sensor placement method at the interface for rapid detection and response based on various leaking scenarios reflecting leaking conditions and environmental conditions of the chemical handling process. In this study, COMSOL was used to analyze the actual accident scenarios by applying the medium parameter to the case of chemical leaks. Based on the accident scenarios, the objective function is selected so that the velocity of each robot is calculated by attaching importance to each item of sensor detection probability, sensing time and sensing scenario number. We also confirmed the feasibility of this method of reliability analysis for unexpected leak accidents. Based on the above results, it is expected that it will be helpful to trace back the leakage source based on the concentration data of the portable sensor to be applied later.

• Journal of the Korea Convergence Society
• /
• v.11 no.9
• /
• pp.1-6
• /
• 2020

In this study, we propose an association inference model based on data bias optimization for road hazard detection. This is a mining model based on association analysis to collect user's personal characteristics and surrounding environment data and provide traffic accident prevention services. This creates transaction data composed of various context variables. Based on the generated information, a meaningful correlation of variables in each transaction is derived through correlation pattern analysis. Considering the bias of classified categorical data, pruning is performed with optimized support and reliability values. Based on the extracted high-level association rules, a risk detection model for personal characteristics and driving road conditions is provided to users. This enables traffic services that overcome the data bias problem and prevent potential road accidents by considering the association between data. In the performance evaluation, the proposed method is excellently evaluated as 0.778 in accuracy and 0.743 in the Kappa coefficient.