학술논문

Dynamic system stochastic identification mixed with video processing: Validation on a real case
Document Type
Conference
Source
2016 11th France-Japan & 9th Europe-Asia Congress on Mechatronics (MECATRONICS) /17th International Conference on Research and Education in Mechatronics (REM) Mechatronics (MECATRONICS) /17th International Conference on Research and Education in Mechatronics (REM), 2016 11th France-Japan & 9th Europe-Asia Congress on. :146-151 Jun, 2016
Subject
Computing and Processing
Engineering Profession
Robotics and Control Systems
Signal Processing and Analysis
Kalman filters
Mathematical model
Solid modeling
Cameras
Two dimensional displays
Covariance matrices
Real-time systems
Dynamic systems
stochastic identification
computer vision
Language
Abstract
Nowadays, visual information is everyday more present, cameras are more and more fast, small and accurate. Improvements in computer vision enable to go further: to consider cameras as smart sensors. In this framework, this article propose to mix complex mechanical models with video processing in order to estimate invisible parameter and to enable the camera to forecast in the video, with a dynamic meaning. The concept is developed on particular cases of objects moving in a scene where the scenario is known. This application is a proof of concept on a case study of a 2D scenario of a ruler sliding on a table. The aim is to prove the feasibility on a real case with real time architecture.