Identifying poses of safe and productive masons using machine learning

Alwasel, Abdullatif; Sabet, Ali Reza; Nahangi, Mohammad; Haas, Carl T.; Abdel‐Rahman, Eihab

doi:10.1016/j.autcon.2017.09.022

Cited by 42 publications

(19 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among the three considered SVM kernel functions, linear and polynomial ones were found to perform classification with high accuracy (92%), similarly to our results. Regarding the Gaussian kernel function, they did not find a high value of accuracy (89%) [52]. Conversely, we obtained a high accuracy with such a kernel, especially in the LL task with the 2 kg load.…”

Section: Discussioncontrasting

confidence: 70%

“…Nevertheless, when considering the level of accuracy and the high values of FP and FN, a small dataset is not an effective choice (see Figures 4 and 5). Furthermore, when considering the TL-l dataset, the processing time result was lower than the result that was obtained in Alwasel et al study [52].…”

Section: Discussioncontrasting

confidence: 70%

“…In the RL task with the 5 kg load, we obtained a 100% TPR value and a 97.4% TNR value (Table 4). In the study of Alwasel et al, a high accuracy was reached in distinguishing between safe and unsafe postures assumed by expert and inexpert masonry workers [52]. They implemented an SVM that was fed with inertial data of the whole kinematic chain.…”

Section: Discussionmentioning

confidence: 99%

“…In industrial environments, SVMs have been applied to classify and recognize the ergonomic posture of movement patterns [40]. In the last few years, activity recognition has usually been based on machine learning algorithms that are fed with inertial sensor data [38,52]. Alwasel et al proposed and validated an SVM algorithm that classified masonry workers' poses as a function of their experience, obtaining a high rate of accuracy 92.11% [52].…”

Section: Introductionmentioning

confidence: 99%

“…In the last few years, activity recognition has usually been based on machine learning algorithms that are fed with inertial sensor data [38,52]. Alwasel et al proposed and validated an SVM algorithm that classified masonry workers' poses as a function of their experience, obtaining a high rate of accuracy 92.11% [52]. In 2016, Ryu and colleagues tested the action recognition of masonry workers by using supervised learning algorithms, such as an SVM, a k-nearest neighbor (kNN) algorithm, and a neural network [53].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Measuring Biomechanical Risk in Lifting Load Tasks Through Wearable System and Machine-Learning Approach

Conforti

Mileti

Prete

et al. 2020

Sensors

View full text Add to dashboard Cite

Ergonomics evaluation through measurements of biomechanical parameters in real time has a great potential in reducing non-fatal occupational injuries, such as work-related musculoskeletal disorders. Assuming a correct posture guarantees the avoidance of high stress on the back and on the lower extremities, while an incorrect posture increases spinal stress. Here, we propose a solution for the recognition of postural patterns through wearable sensors and machine-learning algorithms fed with kinematic data. Twenty-six healthy subjects equipped with eight wireless inertial measurement units (IMUs) performed manual material handling tasks, such as lifting and releasing small loads, with two postural patterns: correctly and incorrectly. Measurements of kinematic parameters, such as the range of motion of lower limb and lumbosacral joints, along with the displacement of the trunk with respect to the pelvis, were estimated from IMU measurements through a biomechanical model. Statistical differences were found for all kinematic parameters between the correct and the incorrect postures (p < 0.01). Moreover, with the weight increase of load in the lifting task, changes in hip and trunk kinematics were observed (p < 0.01). To automatically identify the two postures, a supervised machine-learning algorithm, a support vector machine, was trained, and an accuracy of 99.4% (specificity of 100%) was reached by using the measurements of all kinematic parameters as features. Meanwhile, an accuracy of 76.9% (specificity of 76.9%) was reached by using the measurements of kinematic parameters related to the trunk body segment.

show abstract

Section: Discussioncontrasting

confidence: 70%

Section: Discussioncontrasting

confidence: 70%

Section: Discussionmentioning

confidence: 99%