• International Journal of Advanced Culture Technology
• /
• 제11권1호
• /
• pp.370-375
• /
• 2023

As we enter a super-aged society, studies are being conducted to reduce complications and deaths caused by falls in elderly adults. Research is being conducted on interventions for preventing falls in the elderly, wearable devices for detecting falls, and methods for improving the performance of fall detection algorithms. Wearable devices for detecting falls of the elderly generally use gyro sensors. In addition, to improve the performance of the fall detection algorithm, an artificial intelligence algorithm is applied to the x, y, z coordinate data collected from the gyro sensor. In this paper, we develop a wearable device that uses a gyro sensor, body temperature, and heart rate sensor for health management as well as fall detection for the elderly. In addition, we develop a fall detection and health management system that works with wearable devices and a guardian's mobile app to improve the performance of the fall detection algorithm and provide health information to guardians.

• Journal of information and communication convergence engineering
• /
• 제11권3호
• /
• pp.199-206
• /
• 2013

An emergency monitoring system for the elderly, which uses acceleration data measured with an accelerometer, angular velocity data measured with a gyroscope, and heart rate measured with an electrocardiogram, is proposed. The proposed fall detection algorithm uses multiple parameter combinations in which all parameters, calculated using tri-axial accelerations and bi-axial angular velocities, are above a certain threshold within a time period. Further, we propose an emergency detection algorithm that monitors the movements of the fallen elderly person, after a fall is detected. The results show that the proposed algorithms can distinguish various types of falls from activities of daily living with 100% sensitivity and 98.75% specificity. In addition, when falls are detected, the emergency detection rate is 100%. This suggests that the presented fall and emergency detection method provides an effective automatic fall detection and emergency alarm system. The proposed algorithms are simple enough to be implemented into an embedded system such as 8051-based microcontroller with 128 kbyte ROM.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.226-228
• /
• 2021

구글사에서 출시된 ML Kit API의 Pose detection를 사용한 영상기반 낙상 알고리즘을 제안한다. Pose detection 알고리듬을 사용하여 추출된 신체의 33개의 3차원 특징점을 활용하여 낙상을 인식한다. 추출된 특징점을 분석하여 낙상을 인식하는 알고리듬은 k-NN을 사용한다. 영상의 크기와 영상내의 인체의 크기에 영향을 받지 않도록 정규화과정을 거치며 특징점들의 상대적인 움직임을 분석하여 낙상을 인식한다. 본 실험을 위해 사용한 13개의 테스트 영상중 13개의 영상에서 낙상을 인식하여 100%의 성공률을 보였다.

• 한국위성정보통신학회논문지
• /
• 제12권2호
• /
• pp.56-60
• /
• 2017

본 논문은 영상처리 기반의 낙상 감지 알고리즘의 설계 및 구현에 관한 내용을 기술한다. 영상처리 기반의 낙상 감지 알고리즘은 카메라로 획득한 입력 영상을 그레이 스케일 변환 후 배경차분과 이진화를 통해 객체를 분리하고, 라벨링을 통해 인체를 인식한다. 인식된 인체는 출력 영상으로 확인이 가능하며 낙상을 감지하게 되면 알람이 발생한다. 컴퓨터 시뮬레이션을 통하여 제안한 알고리즘을 실험한 결과 90%의 검출율을 보여주었다. DSP 영상처리 보드에 구현한 시제품 시험을 통하여 기능을 검증함으로서 실용화 가능성을 확인하였다.

• 한국정보통신학회논문지
• /
• 제18권9호
• /
• pp.2101-2108
• /
• 2014

영상 정보를 이용한 자동 낙상 감지 알고리즘을 제안한다. 자동으로 낙상을 감지하기 위한 낙상 특징 파라미터를 추출하기 위해서 영상정보를 광류 방식에 적용하여 움직임 값들을 추출하고 이 움직임 값들에 대한 전체적인 변화의 정도와 기울기, 중심점을 주성분 분석 방법으로 계산한다. 계산된 고유값과 고유 벡터를 사용하여 6가지 낙상 특징 파라미터를 정의한다. 이 낙상특징파라미터가 미리 정해둔 임계값을 초과하는 경우를 낙상으로 판단하는 단순 임계치 방법과 낙상특징파라미터를 은닉 마르코프 모델(Hidden Markov Model; HMM)에 적용시켜 낙상을 판단하는 방법과 단순임계치와 은닉 마르코프 모델을 결합한 낙상 감지 방법을 제안하고 그 결과를 비교 및 분석한다. 단순 임계치와 은닉 마르코프 모델을 결합한 방법은 단순임계치 방법으로 낙상 가능한 행동들을 결정하고 이 결정된 낙상 행동들만을 은닉 마르코프 모델을 적용하여 낙상을 감지한다. 이 방법은 계산량을 줄이면서 감지 정확도를 유지하는 결과를 보인다.

• 한국IT서비스학회지
• /
• 제12권4호
• /
• pp.435-447
• /
• 2013

If we use a smartphone to analyze and detect falling, it is a huge advantage that the person with a sensor attached to one's body is free from awareness of difference and limitation of space, unlike attaching sensors on certain fixed areas. In this paper, we suggested effective posture analysis of smartphone users, and fall detecting system. Suggested algorithm enables to detect falling accurately by using the fact that instantaneous change of acceleration sensor is different according to user's posture. Since mobile applications working on smart phones are low in compatibility according to mobile platforms, it is a constraint that new development is needed which is suitable for sensor equipment's characteristics. In this paper, we suggested posture analysis algorithm using smartphones to solve the problems related to user's inconvenience and limitation of development according to sensor equipment's characteristics. Also, we developed fall detection system with the suggested algorithm, using hybrid mobile application which is not limited to platform.

• 한국정보전자통신기술학회논문지
• /
• 제9권1호
• /
• pp.38-44
• /
• 2016

본 논문에서는 움직임 벡터를 이용한 낙상 감지 시스템에 관해 기술한다. 두드러진 움직임을 위한 가중치 차영상 기법, 움직임 벡터를 이용하여 인간이라고 판단되는 블랍을 검출하고, 추출된 움직임 벡터를 이용하여 낙상 여부를 판단한다. 기존의 영상 기반 낙상 감지 시스템의 경우 특정 방향으로 낙상이 발생하는 경우에만 낙상 감지에 성공하였지만 제안 시스템의 경우 다양한 각도에서 낙상이 발생하여도 상황 판단이 가능하다는 장점이 있다. 실험을 위해서 150개의 상황을 연출하였으며, 약 85% ~ 97.1% 낙상 상황 판단 성공률을 보였다.

• 센서학회지
• /
• 제20권6호
• /
• pp.420-427
• /
• 2011

The population of elderly people increases rapidly as our society moves towards the aged one. Healthcare for the elderly becomes an important issue and falling down is one of the critical problems although not well recognized. In this study, a fall detection system was developed using a 3-axis accelerometer. Analyzing fall patterns, we took into account the degree of impact, posture angle, the repetitions of similar movements and the activities after a potential fall and proposed an algorithm of fall detection. Information of the fall sensor was sent to a remote healthcare server through the wireless networks of Zigbee and WLAN. Our system was designed to monitor multiples users. 12 persons participated in experiment and each one performed 24 different movements. Our proposed algorithm was compared with other reported ones. Our method produced the excellent results having a sensitivity of 96.4 % and a specificity of 100 % whereas other methods had a sensitivity range between 87.5 % and 94.8 % and a specificity range between 63.5 % and 83.3 %.

• IEIE Transactions on Smart Processing and Computing
• /
• 제5권5호
• /
• pp.327-330
• /
• 2016

Currently, the world's elderly population continues to grow at a dramatic rate. As the number of senior citizens increases, detection of someone falling has attracted increasing attention for visual surveillance systems. This paper presents a novel fall-detection algorithm using motion estimation and an integrated spatiotemporal energy map of the object region. The proposed method first extracts a human region using a background subtraction method. Next, we applied an optical flow algorithm to estimate motion vectors, and an energy map is generated by accumulating the detected human region for a certain period of time. We can then detect a fall using k-nearest neighbor (kNN) classification with the previously estimated motion information and energy map. The experimental results show that the proposed algorithm can effectively detect someone falling in any direction, including at an angle parallel to the camera's optical axis.

• 대한의용생체공학회:의공학회지
• /
• 제44권6호
• /
• pp.384-391
• /
• 2023

Among the elderly, fatal injuries and deaths are significantly attributed to falls. Therefore, a pre-impact fall detection system is necessary for injury prevention. In this study, a robust threshold-based algorithm was proposed for pre-impact fall detection, reducing false positives in highly dynamic daily-living movements. The algorithm was validated using public datasets (KFall and FARSEEING) that include the real-world elderly fall. A 6-axis IMU sensor (Movella Dot, Movella, Netherlands) was attached to S2 of 20 healthy adults (aged 22.0±1.9years, height 164.9±5.9cm, weight 61.4±17.1kg) to measure 14 activities of daily living and 11 fall movements at a sampling frequency of 60Hz. A 5Hz low-pass filter was applied to the IMU data to remove high-frequency noise. Sum vector magnitude of acceleration and angular velocity, roll, pitch, and vertical velocity were extracted as feature vector. The proposed algorithm showed an accuracy 98.3%, a sensitivity 100%, a specificity 97.0%, and an average lead-time 311±99ms with our experimental data. When evaluated using the KFall public dataset, an accuracy in adult data improved to 99.5% compared to recent studies, and for the elderly data, a specificity of 100% was achieved. When evaluated using FARSEEING real-world elderly fall data without separate segmentation, it showed a sensitivity of 71.4% (5/7).