Desarrollo y aplicación de un sistema basado en visión artificial para la evaluación ergonómica en diferentes puestos de trabajo

Abstract

The main objective of this degree work is the development and application of a system that performs an ergonomic evaluation with an artificial vision to improve the body postures of workers in different jobs, avoiding irreversible damage to the body, for which software will be developed for the recognition of body segments in the human body. The study will begin with a general description of all the activities carried out in the carpentry area of the cutting and assembly workstation, identifying the existence of high, medium, and low ergonomic risks. The project will be developed through the recognition of people using a webcam, which will process the images captured by means of Python as a programming language and neural networks to allow the recognition of the human body in which place it is located and thus carrying out the segmentation of the body, thus obtaining the measurements of the human body of any person regardless of the height of the worker. The methods used are OWAS, OpenCV, OpenPose, and Python as a programming language. It was possible to identify which activities in the workstations represent the greatest ergonomic problem when lifting loads. The main consequences of ergonomic origin to which the worker is exposed are cervical pain, herniated disc, back pain, and musculoskeletal injuries. It also proposes the respective preventive measures aimed at preventing ergonomic risks, as well as a health check for workers. The aim of this project is to reduce the level of ergonomic risk necessary to prevent ergonomic injuries and at the same time to provide the necessary information for workers to prevent injuries in the short or long term. For more optimal visualization, a gamer computer is necessary.