Camera calibration using detection and neural networks

• Autor(a) principal: Pedra, André Vitor Bosisio Moura
• Data de Publicação: 2013
• Outros Autores: Mendonça, Marcio, Finocchio, Marco Antonio Ferreira, De Arruda, Lúcia Valéria Ramos, Castanho, José Eduardo Cogo [UNESP]
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.3182/20130522-3-BR-4036.00077; http://hdl.handle.net/11449/227192
• topic: Camera calibration; Computer vision; Harris corner extraction; Neural networks

Several applications use robotic vision, such as a robot navigating through an unknown surrounding, can use vision as main navigate sensor. This paper focuses on studying camera calibration via stereo vision by means of neural network. A neurocalibration method is proposed based on the neural networks ability to learn nonlinear relationship among a two and three dimension coordinate systems and also its information generalization skill. The data used to train neural network mapping are generated from a calibration grid point obtained through the use of Harris edge detection algorithm. The experimental results indicated that the neurocalibration method is feasible and efficient. © 2013 IFAC.