• 한국정보통신학회논문지
• /
• 제25권11호
• /
• pp.1649-1654
• /
• 2021

본 논문에서는 고령자의 낙상상황을 감지할 수 있는 텐서플로우 장단기 메모리 기반 낙상감지 시스템의 정규화에 대하여 소개한다. 낙상감지는 고령자의 몸에 부착한 3축 가속도 센서 데이터를 사용하며, 총 7가지의 행동 패턴들에 대하여 학습하며, 각각 4가지는 일상생활에서 일어나는 패턴이고, 나머지 3가지는 낙상에 대한 패턴이다. 학습시에는 손실함수(loss function)를 효과적으로 줄이기 위하여 정규화 과정을 진행하며, 정규화 과정은 데이터에 대하여 최대최소 정규화, 손실함수에 대하여 L2 정규화 과정을 진행한다. 3축 가속도 센서를 이용하여 구한 다양한 파라미터에 대하여 정규화 과정의 최적의 조건을 제시한다. 낙상 검출율면에서 SVM을 이용하고 정규화 127과 정규화율 λ 0.00015일 때 Sensitivity 98.4%, Specificity 94.8%, Accuracy 96.9%로 가장 좋은 모습을 보였다.

• 한국산학기술학회논문지
• /
• 제22권2호
• /
• pp.127-133
• /
• 2021

낙상 판단을 위한 최근 발표되는 연구는 RNN(Recurrent Neural Network)을 이용한 낙상 동작 특징 분석과 동작 분류에 집중되어 있다. 웨어러블 센서를 기반으로 한 접근 방식은 높은 탐지율을 제공하나 사용자의 착용 불편으로 보편화 되지 못했고 최근 영상이나 이미지 기반에 딥러닝 접근방식을 이용한 낙상 감지방법이 소개 되었다. 본 논문은 2D RGB 저가 카메라에서 얻은 영상을 PoseNet을 이용해 추출한 인체 골격 키포인트(Keypoints) 정보로 머리와 어깨의 키포인트들의 위치와 위치 변화 가속도를 추정함으로써 낙상 판단의 정확도를 높이기 위한 감지 방법을 연구하였다. 특히 낙상 후 자세 특징 추출을 기반으로 Convolutional Neural Networks 중 Gated Recurrent Unit 기법을 사용하는 비전 기반 낙상 감지 솔루션을 제안한다. 인체 골격 특징 추출을 위해 공개 데이터 세트를 사용하였고, 동작분류 정확도를 높이는 기법으로 코, 좌우 눈 그리고 양쪽 귀를 포함하는 머리와 어깨를 하나의 세그먼트로 하는 특징 추출 방법을 적용해, 세그먼트의 하강 속도와 17개의 인체 골격 키포인트가 구성하는 바운딩 박스(Bounding Box)의 높이 대 폭의 비율을 융합하여 실험을 하였다. 제안한 방법은 기존 원시골격 데이터 사용 기법보다 낙상 탐지에 보다 효과적이며 실험환경에서 약 99.8%의 성공률을 보였다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제14권6호
• /
• pp.706-711
• /
• 2014

A low power 6-bit flash ADC that uses an input voltage range detection algorithm is described. An input voltage level detector circuit has been designed to overcome the disadvantages of the flash ADC which consume most of the dynamic power dissipation due to comparators array. In this work, four digital input voltage range detectors are employed and each input voltage range detector generates the specific clock signal only if the input voltage falls between two adjacent reference voltages applied to the detector. The specific clock signal generated by the detector is applied to turn the corresponding latched comparators on and the rest of the comparators off. This ADC consumes 68.82 mW with a single power supply of 1.2V and achieves 4.3 effective number of bits for input frequency up to 1 MHz at 500 MS/s. Therefore it results in 4.6 pJ/step of Figure of Merit (FoM). The chip is fabricated in 0.13-um CMOS process.

• 대한전자공학회논문지
• /
• 제24권5호
• /
• pp.733-739
• /
• 1987

In this paper, variable threshold detection with weighted pulse code modulation-encoded signals transmitted over Gaussian channels has been investigated. Each bit in the \ulcornerlaw PCM word is weighted according to its significance in the transmitter. It the output falls into the erasure zone, the regenerated sample replaced by interpolation or prediction. To overall system signal to noise ratio for BPSK/PCM speech signals of this technique has been found. When the input signal level was -17 db, the gains in overall signal s/n compared to weighted PCM and variable threshold detection were 5 db and 3 db, respectively. Computer simulation was performed generating signals by computer. The simulation was in resonable agreement with our theoretical prediction.

• Journal of Trauma and Injury
• /
• 제19권1호
• /
• pp.54-58
• /
• 2006

Purpose: The committee on trauma of the american college of surgeons, in its manual resources for optimal care of the injured patients involved in falls from less than 20 feet need not be taken to trauma centers. Because triage criteria dictate less urgency for low-level falls, this classification scheme has demerits for early detection and treatment of serious problems in the emergency room. Methods: A prospective analysis was conducted of 182 patients treated for fall-related trauma from June 2003 to March 2004. Falls were classified as group A (<3 m), group B (${\geq}3m$, <6 m), and group C (${\geq}6m$). Collected data included the patient's age, gender, site and height of fall, surface fallen upon, body area of first impact, body regions of injuries, Glasgow Coma Scale (GCS), Revised Trauma Score (RTS), and Injury Severity Score (ISS). Results: The 182 patients were classified as group A (105) 57.7%, group B (61) 33.5%, and group C (16) 8.8%. There was a weak positive correlation between the height of fall and the patients' ISS in the three groups (p<0.001). There were significant differences in GCS (p=0.017), RTS (p=0.034), and ISS (p=0.007) between group A and B. In cases that the head was the initial impact area of the body, the GCS (p<0.001) and the RTS (p=0.002) were lower, but the ISS (p<0.001) was higher than it was for other type of injuries. Hard surfaces as an impact surface type, had an influence on the GCS (p<0.001) and the ISS (p=0.025). Conclusion: To simply categorize patients who fall over 6 meters as severely injured patients doesn't have much meaning, and though patients may have fallen less than 6 meters, they should be categorized by using the dynamics (impact surface type, initial body - impact area) of their fall.

• International Journal of Internet, Broadcasting and Communication
• /
• 제12권4호
• /
• pp.83-92
• /
• 2020

A recent study to determine the fall is focused on analyzing fall motions using a recurrent neural network (RNN), and uses a deep learning approach to get good results for detecting human poses in 2D from a mono color image. In this paper, we investigated the improved detection method to estimate the position of the head and shoulder key points and the acceleration of position change using the skeletal key points information extracted using PoseNet from the image obtained from the 2D RGB low-cost camera, and to increase the accuracy of the fall judgment. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion analysis method and on the velocity of human body skeleton key points change as well as the ratio change of body bounding box's width and height. The public data set was used to extract human skeletal features and to train deep learning, GRU, 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 the conventional primitive skeletal data use method.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.391-393
• /
• 2021

본 논문에서는 일상생활에서의 고령자에게 나타날 수 있는 낙상상황을 감지할 수 있는 텐서플로우를 이용한 장단기 메모리 기반 낙상감지 시스템에 대하여 소개한다. 낙상감지를 위해서 3축 가속도 센서 데이터를 이용하고, 이를 처리하여 다양한 파라미터화하며 일상생활 패턴 4가지, 낙상상황 패턴 3가지로 분류한다. 파라미터화한 데이터는 정규화 과정을 따르며, 학습이 진행된다. 학습은 Loss값이 0.5 이하가 될 때까지 진행된다. 각각의 파라미터인 θ, SVM (Sum Vector Magnitude), GSVM (gravity-weight SVM)에 대하여 결과를 산출한다. 가장 좋은 결과는 GSVM으로 Sensitivity 98.75%, Specificity 99.68%, Accuracy 99.28%로 가장 좋은 결과를 보였다.

• 센서학회지
• /
• 제33권1호
• /
• pp.18-23
• /
• 2024

Humanoid robots, designed to interact in human environments, require stable mobility to ensure safety. When a humanoid robot falls, it causes damage, breakdown, and potential harm to the robot. Therefore, fall detection is critical to preventing the robot from falling. Prevention of falling of a humanoid robot requires an operator controlling a crane. For efficient and safe walking control experiments, a system that can replace a crane operator is needed. To replace such a crane operator, it is essential to detect the falling conditions of humanoid robots. In this study, we propose falling detection methods using Convolution Neural Network (CNN) model. The image data of a humanoid robot are collected from various angles and environments. A large amount of data is collected by dividing video data into frames per second, and data augmentation techniques are used. The effectiveness of the proposed CNN model is verified by the experiments with the humanoid robot MAX-E1.

• 동서간호학연구지
• /
• 제28권2호
• /
• pp.122-131
• /
• 2022

Purpose: This study aimed to identify the factors related to fear of falling (FOF) in different age groups from community-dwelling mid to late-adults. Methods: To identify the factors related to FOF, data of 162,684 adults over 45 years of age from 2019 Community Health Survey was analyzed using logistic regression with complex samples. Results: Factors related to FOF found in all age groups were sex, previous experience of falls, physical activity levels over moderate intensity, subjective health status, number of chronic diseases, stress, depression, and cognitive decline. In the 45-64 age group, the FOF was significantly higher in the groups of low education level and low monthly household income. In the 65-74 and over 75 age groups, the FOF was significantly higher in the groups of not living with spouse and walking not practiced. Conclusion: We suggests that understanding of risk factors and early detection of fall risk patients in each age group are necessary to establish and apply tailored fall prevention programs for prevention and management of the FOF in community-dwelling mid to late-adults.

• 한국산업정보학회논문지
• /
• 제18권4호
• /
• pp.19-24
• /
• 2013

고령화 사회에서는 노인의 신체적 취약성으로 인한 안전문제가 사회문제로 대두되고 있다. 판단, 상황대처 능력이 떨어진 노인은 체력과 균형감각 저하로 인하여 잦은 낙상을 경험한다. 낙상은 자칫 인명피해 및 골절, 유조직 손상 등을 유발 할 수 있으므로 빠른 응급대처가 필요하다. 따라서 본 논문에서는 사용자의 허리에 부착하여 일상적인 움직임에 대한 가속도의 변화 및 낙상이 발생하였을 때의 가속도의 변화를 측정하였다. 측정한 값을 이용하여 낙상 감지 시스템을 구현하였으며, 여러 가지 낙상 상황을 가정하여 낙상 검출 여부를 판별 하였다.