• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.93-94
• /
• 2022

최근 고령화 사회가 심각한 사회적 문제로 급부상하고 있으며, 이에 병원을 찾아 입원하는 비중이 이전에 비하여 높아지고 있다. 거동이 불편하거나 근력이 부족한 환자의 경우 스스로 거동할 능력이 다소 떨어지며, 낙상사고가 발생하면 부상 혹은 치명적일 경우 사망으로 이어질 수 있다. 하지만, 이들을 보살피는 간호 인력만으로 병원 내 모든 낙상사고를 파악하기에는 한계가 있다. 또한, 환자들의 낙상 탐지에 관한 연구는 지속해서 수행되어왔지만, 병원 내부환경에서의 낙상 탐지 연구는 부족하다. 이에 본 논문에서는 병원 내부환경에서 낙상을 탐지하기 위해 실제 병실에서 수집한 데이터로 YOLOv5 모델을 학습하여 환자 낙상 탐지모델을 구축 및 평가하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.491-500
• /
• 2021

In this study, we propose a method for detecting fall behavior using MS Kinect v2 RGBD Camera-based Human-Skeleton Keypoints and a 2-Stacked Bi-LSTM model. In previous studies, skeletal information was extracted from RGB images using a deep learning model such as OpenPose, and then recognition was performed using a recurrent neural network model such as LSTM and GRU. The proposed method receives skeletal information directly from the camera, extracts 2 time-series features of acceleration and distance, and then recognizes the fall behavior using the 2-Stacked Bi-LSTM model. The central joint was obtained for the major skeletons such as the shoulder, spine, and pelvis, and the movement acceleration and distance from the floor were proposed as features of the central joint. The extracted features were compared with models such as Stacked LSTM and Bi-LSTM, and improved detection performance compared to existing studies such as GRU and LSTM was demonstrated through experiments.

• Journal of Korea Society of Industrial Information Systems
• /
• v.21 no.2
• /
• pp.31-37
• /
• 2016

Falling situations are extremely critical events for the elderly person who requires timely and adequate emergency service. For the case of emergency, the information of falling and its direction can be used as an important information for the first aid treatment of the injured person. In this paper, a falling detection system which can pinpoint the falling event with the falling direction is implemented. In order to detect the fall situation, a single gyroscope (MPU-6050) is used in the developed system. The fall detection algorithm that can classify 8 different fall directions such as front, back, left, right and in between falls is proposed. The direction of the fall is decided by examining the acceleration values of X and Y directions of the sensor. It is shown that the proposed algorithm successfully detects the falling event and the falling direction with probability of 97% for a selected value of acceleration threshold.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.13-16
• /
• 2008

본 논문은 가중퍼지소속함수 기반신경망(neural network with weighted fuzzy membership functions, NEWFM)기반의 자동 특징 추출기법을 사용하여 인체의 세 방향에서 발생하는 가속도 값으로부터 낙상을 탐지하는 방안을 제시하고 있다. 10명의 피검자로부터 8가지 시나리오로 낙상/비낙상 데이터 800개를 수집하고 웨이블릿 변환(wavelet transform, WT)을 통해 추출한 계수중 비중복면적 분산법에 의해 중요도가 가장 낮은 특징입력을 하나씩 제거하면서 최소의 특징 입력을 선택하였다. 특징입력으로는 가속도 값을 웨이블릿 변환한 11개의 d4계수들 중 비중복면적 분산법에 의해서 중요도가 가장 높은 5개의 계수가 사용되었고, 이들 특징입력을 통해 93%의 전체 분류율을 나타내었다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.127-133
• /
• 2021

A recent study of people physically falling focused on analyzing the motions of the falls using a recurrent neural network (RNN) and a deep learning approach to get good results from detecting 2D human poses from a single color image. In this paper, we investigate a detection method for estimating the position of the head and shoulder keypoints and the acceleration of positional change using the skeletal keypoints information extracted using PoseNet from an image obtained with a low-cost 2D RGB camera, increasing the accuracy of judgments about the falls. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion-analysis method. A public data set was used to extract human skeletal features, and as a result of an experiment to find a feature extraction method that can achieve high classification accuracy, the proposed method showed a 99.8% success rate in detecting falls more effectively than a conventional, primitive skeletal data-use method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.422-423
• /
• 2023

고령화 사회로 인한 노인의 건강과 안전에 대한 관심이 증가함에 따라 낙상 문제는 더욱 중요해졌다. 기존 연구들은 영상에서 인체의 관절위치를 측정하고 이것만을 활용하여 낙상을 감지했지만, 본 논문에서는 방향과 속력 정보를 추가하여 탐지 능력을 향상시켰다. 실험결과 기존 방식에 비해 향상된 성능을 관찰할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.998-999
• /
• 2023

본 논문에서는 스마트 홈 기반 노약자 응급상황 탐지 및 대응 시스템을 설계하였다. 2 개의 레이다 센서를 활용하여 센서 데이터를 분석하고 분류하며, 사용자의 상태를 취침, 외출, 응급상황 총 3 가지 경우로 식별한다. AWS 서버의 데이터베이스를 통해 응급상황 및 낙상 감지 이력을 축적하여 맞춤형 서비스를 제공한다. 어플리케이션을 통해 응급상황 자동 신고 접수와 센서 오작동시 자동 신고 접수 수동 취소 기능을 제공하는 응급상황 탐지 및 대응 시스템을 소개한다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.6
• /
• pp.23-31
• /
• 2022

Due to the recent outbreak of COVID-19 and an aging population and an increase in single-person households, the amount of time that household members spend doing various activities at home has increased significantly. In this study, we propose an algorithm for detecting anomalies in members of single-person households, including the elderly, based on the results of human movement and fall detection using an image sensor algorithm through home CCTV, an activity sensor algorithm using an acceleration sensor built into a smartphone, and a 2D LiDAR sensor-based LiDAR sensor algorithm. However, each single sensor-based algorithm has a disadvantage in that it is difficult to detect anomalies in a specific situation due to the limitations of the sensor. Accordingly, rather than using only a single sensor-based algorithm, we developed a fusion method that combines each algorithm to detect anomalies in various situations. We evaluated the performance of algorithms through the data collected by each sensor, and show that even in situations where only one algorithm cannot be used to detect accurate anomaly event through certain scenarios we can complement each other to efficiently detect accurate anomaly event.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.150-159
• /
• 2022

This paper presents an implementation of detecting whether a helmet is worn and there is a fall accident through individual image analysis in real-time from extracting the image objects of several workers active in the industrial field. In order to detect image objects of workers, YOLO, a deep learning-based computer vision model, was used, and for whether a helmet is worn or not, the extracted images with 5,000 different helmet learning data images were applied. For whether a fall accident occurred, the position of the head was checked using the Pose real-time body tracking algorithm of Mediapipe, and the movement speed was calculated to determine whether the person fell. In addition, to give reliability to the result of a falling accident, a method to infer the posture of an object by obtaining the size of YOLO's bounding box was proposed and implemented. Finally, Telegram API Bot and Firebase DB server were implemented for notification service to administrators.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.211-218
• /
• 2021

A recent spread of personal mobility devices such as electric kickboards has brought about a rapid increase in accident cases. Such vehicles are susceptible to falling accidents due to their low dynamic stability and lack of outer protection chassis. This paper presents the development of an automatic emergency braking system and a safe starting system as driving assistance devices for electric kickboards. The braking system employed artificial intelligence to detect nearby threaening objects. The starting system was developed to disable powder to the motor until when the driver's boarding is confirmed. This study is meaningful in that it proposes the convergence technology of advanced driver assistance systems specialized for personal mobility devices.