Estimating Trunk Angle Kinematics During Lifting Using a Computationally Efficient Computer Vision Method

Abstract

Objective A computer vision method was developed for estimating the trunk flexion angle, angular speed, and angular acceleration by extracting simple features from the moving image during lifting. Background Trunk kinematics is an important risk factor for lower back pain, but is often difficult to measure by practitioners for lifting risk assessments. Methods Mannequins representing a wide range of hand locations for different lifting postures were systematically generated using the University of Michigan 3DSSPP software. A bounding box was drawn tightly around each mannequin and regression models estimated trunk angles. The estimates were validated against human posture data for 216 lifts collected using a laboratory-grade motion capture system and synchronized video recordings. Trunk kinematics, based on bounding box dimensions drawn around the subjects in the video recordings of the lifts, were modeled for consecutive video frames. Results The mean absolute difference between predicted and motion capture measured trunk angles was 14.7°, and there was a significant linear relationship between predicted and measured trunk angles ( R2 = .80, p < .001). The training error for the kinematics model was 2.3°. Conclusion Using simple computer vision-extracted features, the bounding box method indirectly estimated trunk angle and associated kinematics, albeit with limited precision. Application This computer vision method may be implemented on handheld devices such as smartphones to facilitate automatic lifting risk assessments in the workplace.

Design of Active Display Stand Combing the Pressure Sensor and Kinematics Algorithm

“…Most computer monitor pedestals have no adjustment function during operation. Most of the current display brackets are based on weight and realize multiangle adjustment, but the manufacturing cost is high due to their many structures [18][19][20]. Due to the inappropriate cooperation between people and devices, there are often many redundant connecting rods, and automatic adjustment of the connecting rods cannot be realized during operation.…”

Enhancing the Revised NIOSH Lifting Equation using ComputerVision

“…Trunk kinematics estimation. After classifying the postures for each BLA, trunk angle, speed, and acceleration were estimated for each lift using algorithms developed by Greene et al (2020). Features including bounding box height (BH) and width (BW), hand horizontal location (H) and vertical location (V) were extracted from the 3DSSPP simulated postures to estimate trunk flexion angle (T) at the lifting instance.…”

“…Advances in computing power has also facilitated the expansion of deep learning based methods for analyzing human activity for classifying occupational lifting tasks (Mehrizi et al, 2019). Leveraging computer vision, our research group developed a series of computationally efficient lifting analysis methods including posture classification (Greene et al, 2019), trunk kinematics estimation (Greene et al, 2020), and lift load assessment (Li et al, 2020). Additionally, we have created a lifting monitoring algorithm that continuously analyzes manual lifting tasks in videos and automatically measures the input variables for the revised NIOSH lifting equation or RNLE (Wang et al, 2019).…”