• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.2
• /
• pp.336-342
• /
• 2015

Fall-related injuries are the most common cause of accidental death for the elderly and the most frequent work-related injuries in construction sites. Due to the growing popularity of smartphones, there has been a number of research work related to the use of sensors embedded in the smartphone for fall detection. Falls can be detected easily by measuring the magnitude and direction of acceleration vectors. In general, the direction of the acceleration vector does not show the object movement, but the velocity vector directly indicates the tangential direction in which the object is moving. In this paper, we proposed a new method for computing the fall direction based on the characteristics of the velocity vector extracted from the accelerometer.

• Journal of Creative Information Culture
• /
• v.7 no.4
• /
• pp.235-242
• /
• 2021

Recently, many human activity recognition(HAR) researches using smartphone sensor data have been studied. HAR can be utilized in various fields, such as life pattern analysis, exercise measurement, and dangerous situation detection. However researches have been focused on recognition of basic human behaviors or efficient battery use. In this paper, exercising activities performed indoors and outdoors were defined and recognized. Data collection and pre-processing is performed to recognize the defined activities by SVM, random forest and gradient boosting model. In addition, the recognition result is determined based on voting class approach for accuracy and stable performance. As a result, the proposed activities were recognized with high accuracy and in particular, similar types of indoor and outdoor exercising activities were correctly classified.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.2
• /
• pp.397-407
• /
• 2015

In this paper, we designed 1 chip IC for 3-axis gyroscope and 3-axis accelerometer used for various IoT/M2M mobile devices such as smartphone, wearable device and etc. We especially focused on analysis of gyroscope noise and proposed new architecture for removing various noise generated by gyroscope MEMS and IC. Gyroscope, accelerometer and geo-magnetic sensors are usually used to detect user motion or to estimate moving distance, direction and relative position. It is very important element to designing a low noise IC because very small amount of noise may be accumulated and affect the estimated position or direction. We made a mathematical model of a gyroscope sensor, analyzed the frequency characteristics of MEMS and circuit, designed a low noise, compact and low power 1 chip 6-axis inertial sensor IC including 3-axis gyroscope and 3-axis accelerometer. As a result, designed IC has 0.01dps/${\sqrt{Hz}}$ of gyroscope sensor noise density.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.1048-1050
• /
• 2013

Mobile devices or smartphones are rapidly becoming the central computer and communication network device. Recently, smartphones are programmable and come with a growing set of cheap powerful embedded sensors, such as an accelerometer, digital compass, gyroscope, GPS, microphone, and camera. In this paper, we discuss the wireless gas sensing service architectural and develope the gas sensor monitoring services using smartphone and web server.

• Tunnel and Underground Space
• /
• v.28 no.6
• /
• pp.547-554
• /
• 2018

Using crowdsourced sensor data collection technique, it is possible to collect high-density ground vibration data which is difficult to obtain by conventional methods. In this study, we have developed a crowdsourced ground vibration data collection system using MEMS sensors mounted on small electronic devices including smartphones, and implemented client and server based on the proposed infrastructure system design. The system is designed to gather vibration data quickly through Android-based smartphones or fixed devices based on Android Things, minimizing the usage of resource like power usage and data transmission traffic of the hardware.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.5
• /
• pp.73-79
• /
• 2020

As the number of earthquakes gradually increases on the Korean Peninsula, much research has been actively conducted to detect earthquakes quickly and accurately. Because traditional seismic stations are expensive to install and operate, recent research is currently being conducted to detect earthquakes using low-cost MEMS sensors. In this article, we evaluate how a low-cost MEMS acceleration sensor installed in a smartphone can be used to detect earthquakes. To this end, we installed about 280 smartphones at various locations in Korea to collect acceleration data and then assessed the installed sensors' noise floor through PSD calculation. The noise floor computed from PSD determines the magnitude of the earthquake that the installed MEMS acceleration sensors can detect. For the last few months of real operation, we collected acceleration data from 200 smartphones among 280 installed smartphones and then computed their PSDs. Based on our experiments, the MEMS acceleration sensor installed in the smartphone is capable of observing and detecting earthquakes with a magnitude 3.5 or more occurring within 10km from an epic center. During the last several months of operation, the smartphone acceleration sensor recorded an earthquake of magnitude 3.5 in Miryang on December 30, 2019, and it was confirmed as an earthquake using STA/LTA which is a simple earthquake detection algorithm. The earthquake detection system using MEMS acceleration sensors is expected to be able to detect increasing earthquakes more quickly and accurately.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.11
• /
• pp.537-542
• /
• 2017

Aggressive driving is a major cause of car accidents. Previous studies have mainly analyzed young driver's aggressive driving tendency, yet they were only done through pure clustering or classification technique of machine learning. However, since elderly people have different driving habits due to their fragile physical conditions, it is necessary to develop a new method such as enhancing the characteristics of driving data to properly analyze aggressive driving of elderly drivers. In this study, acceleration data collected from a smartphone of a driving vehicle is analyzed by a newly proposed ECA(Enhanced Clustering method for Acceleration data) technique, coupled with a conventional clustering technique (K-means Clustering, Expectation-maximization algorithm). ECA selects high-intensity data among the data of the cluster group detected through K-means and EM in all of the subjects' data and models the characteristic data through the scaled value. Using this method, the aggressive driving data of all youth and elderly experiment participants were collected, unlike the pure clustering method. We further found that the K-means clustering has higher detection efficiency than EM method. Also, the results of K-means clustering demonstrate that a young driver has a driving strength 1.29 times higher than that of an elderly driver. In conclusion, the proposed method of our research is able to detect aggressive driving maneuvers from data of the elderly having low operating intensity. The proposed method is able to construct a customized safe driving system for the elderly driver. In the future, it will be possible to detect abnormal driving conditions and to use the collected data for early warning to drivers.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.4
• /
• pp.53-63
• /
• 2014

Moving context determination is an important issue to be resolved in a mobile computing environment. This paper presents a method for recognizing and classifying a mobile user's moving context by Euclidean distance similarity. In the proposed method, basic data are gathered using Global Positioning System (GPS) and accelerometer sensors, and by using the data, the system decides which moving situation the user is in. The decided situation is one of the four categories: stop, walking, run, and moved by a car. In order to evaluate the effectiveness and feasibility of the proposed scheme, we have implemented applications using several variations of Euclidean distance similarity on the Android system, and measured the accuracies. Experimental results show that the proposed system achieves more than 90% accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.11
• /
• pp.2423-2428
• /
• 2010

The number of motor vehicle registrations in Korea is increasing steadily each year, driven by industry development and economic growth. The number of traffic accidents is also rapidly increasing. Korea has a relatively high number of traffic accidents among OECD member countries, and it ranks among the highest in traffic accident death rates. This death rate is higher compared to death rates as a proportion of the number of traffic accidents in each country. It is very common for drivers to lose consciousness in traffic collisions, which leads to a failure to carry out early emergency measures. In order to prevent such situations as well as hit-and-runs and people left uncared for after traffic accidents, there is a need for motor vehicle black boxes and accident report systems. This study addressed the need for an emergency evacuation system for people injured in traffic accidents and a secondary traffic accident prevention system by developing a motor vehicle emergency situation detection and report system combined with a black box, and materializing it as an actual system.