Abstract

An intelligent method for improving position linearity of position-sensitive detector (PSD), based on support vector machines (SVMs), is developed. The SVM is established based on the structural risk minimization principle rather than minimizing the empirical error commonly implemented in neural networks. SVM can achieve higher generalization performance. Training SVM is equivalent to solving a linearly constrained quadratic programming problem, thus the solution of SVM is always unique and globally optimal. The improving position linearity procedure has been illustrated using a two-dimensional (2D) PSD. It is pointed out that the position linearity of the measuring system with a proper SVM correction is improved by two orders of magnitude in the measurement range.

© 2005 Chinese Optics Letters

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription