Aplicación de la Visión por Computador a la docencia de Fisica
[EN] This thesis presents, develops and validates a new methodology based on video analysis for the lessons related to dynamics during the first Physics courses at Universities. Instead of applying traditional approaches based on comparing the measured trajectory, r(t), with such associated to a det...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Fecha de publicación: | 2015 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | español |
| OAI Identifier: | oai:riunet.upv.es:10251/59470 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/59470 |
| Access Level: | acceso abierto |
| Palabra clave: | Videoanálisis Fotogrametría Video Fotogrametría Prácticas de laboratorio Física Experimentos de laboratorio Laboratorio de Física FISICA APLICADA |
| Sumario: | [EN] This thesis presents, develops and validates a new methodology based on video analysis for the lessons related to dynamics during the first Physics courses at Universities. Instead of applying traditional approaches based on comparing the measured trajectory, r(t), with such associated to a determined model, in this thesis it is proposed to analyze directly the relation among the variables which define the dynamics of the system, i.e. the relation among position, speed and acceleration. In this way, it is possible to compare the differential equation that regulates the dynamic of the movement, providing a different perspective on the problems. Among others, one major advantage of the proposed approach is to ease the explanation of some phenomena because it is only necessary to describe the equation of the dynamics but without the requirement of its integration. This is considerable simpler for the students and lets them analyze other interesting dynamic systems which are not used to be studied during the first courses at University. On the other hand, this approach presents some tools, like the diagrams in thephases space, which ease the graphic interpretation of system evolution. Finally, the equations of the dynamics are quite more sensitive to the deviations between the model and the reality of its integrals. Thus, the students can assess and validate some hypotheses like the linearity or the absence of friction. In order to apply this approach, a computer vision system is developed and adapted to the requirements of a teaching lab of physics, but with the same performance as a research lab. The main limitations of current teaching systems of video analysis are identified, and the main problems related to the configuration of the cameras, their calibration, the automatic digitalization and the accuracy in the movement acquisition are solved. A result of this development is an automatic computer vision system ofhigh precision and interoperability with both cameras of low cost and a more sophisticated performance. The video acquisition system and the movement analysis integrate some specific algorithms for smoothing and numeric derivation. These algorithms are used to calculate accurately speed and acceleration of a movement. The final system (providing a smoother capture of movements) is validated through some experiments. The accuracy in position measuring is about 0.1%, 0.5% for speed and 2% for acceleration, even though using web cameras of very low cost. Nowadays, there is not any teaching lab equipment able to measure continuously and simultaneously these cinematic magnitudes with so high accuracy while being so cheap. In order to show the multitude of possibilities that are offered by this new approach, five practices of dynamics using this video analysis system are designed, covering aspects like the cinematic of the point and the solid, the non linear oscillations, the systems with dynamic balance, the systems with variable mass or the dynamic of rigid solid. |
|---|