Fall-from-Height Detection Using Deep Learning Based on IMU Sensor Data for Accident Prevention at Construction Sites
TL;DR: This study successfully predicted the FFH risk levels and could be helpful to reduce fatal injuries at construction sites and the Conv-LSTM model trained by MAE showed the smallest error and the classification with the predicted peak acceleration showed the best accuracy.
read more
Abstract: Workers at construction sites are prone to fall-from-height (FFH) accidents. The severity of injury can be represented by the acceleration peak value. In the study, a risk prediction against FFH was made using IMU sensor data for accident prevention at construction sites. Fifteen general working movements (NF: non-fall), five low-hazard-fall movements, (LF), and five high-hazard-FFH movements (HF) were performed by twenty male subjects and a dummy. An IMU sensor was attached to the T7 position of the subject to measure the three-axis acceleration and angular velocity. The peak acceleration value, calculated from the IMU data, was 4 g or less in general work movements and 9 g or more in FFHs. Regression analysis was performed by applying various deep learning models, including 1D-CNN, 2D-CNN, LSTM, and Conv-LSTM, to the risk prediction, and then comparing them in terms of their mean absolute error (MAE) and mean squared error (MSE). The FFH risk level was estimated based on the predicted peak acceleration. The Conv-LSTM model trained by MAE showed the smallest error (MAE: 1.36 g), and the classification with the predicted peak acceleration showed the best accuracy (97.6%). This study successfully predicted the FFH risk levels and could be helpful to reduce fatal injuries at construction sites.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Near-Miss Detection Metrics: An Approach to Enable Sensing Technologies for Proactive Construction Safety Management
Filzah Hashmi,Muhammad Usman Hassan,Muhammad Umer Zubair,Khursheed Ahmed,Taha Aziz,Rafiq M. Choudhry +5 more
TL;DR: Near-miss detection metrics for automated reporting and documentation in construction safety management. The study developed quantifiable near-miss definitions, identified significant near-misses, and determined measurable data required for sensor-based detection.
5
Feasibility Analysis of Using Channel State Information (CSI) Acquired from Wi-Fi Routers for Construction Worker Fall Detection
TL;DR: In this paper , the authors investigated the potential of using Channel State Information (CSI) to identify falls among construction workers and showed that the behavior of construction workers is highly correlated with the magnitude of CSI, even in real construction sites, and the CSI-based method for identifying construction worker falls has an accuracy of 99% and can also accurately distinguish between falls and fall-like actions.
Monitoring and evaluating the status and behaviour of construction workers using wearable sensing technologies
Mingzhu Wang,Jiayu Chen,Jie Ma +2 more
4
Computerized Predictive Analysis of Accidents (FOCOS/PRONTOS System): Study of Rare Cases
TL;DR: The Computerized Test for Readiness Evaluation (FOCOS/Prontos System) as a risk assessment tool identified alterations in readiness parameters in 100% of the workers involved in the analyzed cases, demonstrating its predictive capacity in identifying accident risks as mentioned in this paper .
1
Risk Prediction in Daily Activities and Falls based on Deep Learning
Seunghee Lee,Bummo Koo,Sumin Yang,Dongkwon Kim,Youngho Kim +4 more
TL;DR: This study uses deep learning regression models to predict fall risk in the elderly by analyzing 6-axis IMU data from 20 healthy adults performing 14 daily activities and 11 falls, achieving lower error values with data augmentation.
References
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
99K
Recent advances in convolutional neural networks
Jiuxiang Gu,Zhenhua Wang,Jason Kuen,Lianyang Ma,Amir Shahroudy,Bing Shuai,Ting Liu,Xingxing Wang,Gang Wang,Jianfei Cai,Tsuhan Chen +10 more
TL;DR: A broad survey of the recent advances in convolutional neural networks can be found in this article, where the authors discuss the improvements of CNN on different aspects, namely, layer design, activation function, loss function, regularization, optimization and fast computation.
5.7K
A CNN-LSTM-Based Model to Forecast Stock Prices
TL;DR: A forecasting method of stock price based on CNN-LSTM which can provide a reliable stock price forecasting with the highest prediction accuracy and provides practical experience for scholars to study financial time series data is proposed.
A deep learning-based approach for mitigating falls from height with computer vision: Convolutional neural network
Weili Fang,Weili Fang,Botao Zhong,Neng Zhao,Peter E.D. Love,Hanbin Luo,Jiayue Xue,Shuangjie Xu +7 more
TL;DR: It is suggested that proposed computer-vision approach that is developed can be used by site management to automatically identify unsafe behavior and provide feedback to individuals about their likelihood of falls from heights by recognizing unsafe behavior in real-time.
208
Semi-supervised near-miss fall detection for ironworkers with a wearable inertial measurement unit
TL;DR: A semi-supervised learning algorithm (i.e., one-class support vector machine) was implemented for detecting the near-miss falls in this article, and two experiments were conducted for collecting the near miss falls of ironworkers, and these data were used to test developed near miss fall detection approach.
167