An implementation of the partitioned Levenberg-Marquardt algorithm for applications in computer vision.

At several applications of computer vision is necessary to estimate parameters for a specific model which best fits an experimental data set. For these cases, a minimization algorithm might be used and one of the most popular is the Levenberg-Marquardt algorithm. Although several free applies from t...

ver descrição completa

Detalhes bibliográficos
Autores: França, José Alexandre de, França, Maria Bernadete de Morais, Koyama, Marcela Hitomi, Silva, Tiago Polizer da
Formato: artículo
Estado:Versión publicada
Fecha de publicación:2009
País:Brasil
Recursos:Universidade Estadual de Londrina (UEL)
Repositorio:Revista Semina: Ciências Exatas e Tecnológicas (Online)
Idioma:portugués
OAI Identifier:oai:ojs2.ojs.uel.br:article/4734
Acesso em linha:https://ojs.uel.br/revistas/uel/index.php/semexatas/article/view/4734
Access Level:acceso abierto
Palavra-chave:Levenberg-Marquardt algorithm
Monocular Calibration
Newton's.
Software system
Algoritmo Levenberg-Marquardt
Calibração Monocular
Algoritmo de Newton.
Sistema de software
Descrição
Resumo:At several applications of computer vision is necessary to estimate parameters for a specific model which best fits an experimental data set. For these cases, a minimization algorithm might be used and one of the most popular is the Levenberg-Marquardt algorithm. Although several free applies from this algorithm are available, any of them has great features when the resolution of problem has a sparse Jacobian matrix . In this case, it is possible to have a great reduce in the algorithm's complexity. This work presents a Levenberg-Marquardt algorithm implemented in cases which has a sparse Jacobian matrix. To illustrate this algorithm application, the camera calibration with 1D pattern is applied to solve the problem. Empirical results show that this method is able to figure out satisfactorily with few iterations, even with noise presence.