Ergonomic risk assessment application based on computer vision and machine learning

Musculoskeletal disorders (MSDs) account for the majority of work -related non -fatal diseases in industry. In the literature, there are many ergonomic risk assessment methods and software that implements them, from simple checklists to complex assessments to prevent MSD-related diseases. However, in these applications, while angles are calculated automatically, relative questions such as arm grip success, shoulder and arm support are directed to the user with an interface. In this study, a web -based platform has been developed that can simultaneously provide ergonomic risk assessment (ERD) reports for REBA, RULA and OWAS methods with the Region -based Convolutional Neural Network (R -CNN) of MediaPipe machine learning library. The evaluation and comparison algorithm on the platform and the relative questions within the ERD methods will also be answered by the developed application, ensuring consistency and ease of use. With this aspect of the study, it is aimed to fill the gap in the literature. For the validation of the proposed platform, the Object Keypoint Similarity (OKS) test used in pose estimation algorithms was applied. The test was applied to each of the 32 body key points, with an overall average accuracy of 92%. In the other test process, the accuracy of the measured body joint angles was calculated to be used in ERD methods. Each of the 13 body joint angles was compared with the actual baseline angles and an average RMSE (Root Mean Square Error) of 7.7 degrees was obtained. When the RMSE value and OKS result obtained were compared with the current literature, it was determined that the values were consistent.