Abstract

In a structured-light system, lens distortion of the camera and projector is the main source of 3D measurement error. In this Letter, a new approach, to the best of our knowledge, of using deep neural networks to address this problem is proposed. The neural network consists of one input layer, five densely connected hidden layers, and one output layer. A ceramic plate with flatness less than 0.005 mm is used to acquire the training, validation, and test data sets for the network. It is shown that the measurement accuracy can be enhanced to 0.0165 mm in the RMS value by this technique, which is an improvement of 93.52%. It is also verified that the constructed neural network is with satisfactory repeatability.

© 2019 Optical Society of America

Full Article  |  PDF Article
OSA Recommended Articles
Dynamic-range compression scheme for digital hologram using a deep neural network

Tomoyoshi Shimobaba, David Blinder, Michal Makowski, Peter Schelkens, Yota Yamamoto, Ikuo Hoshi, Takashi Nishitsuji, Yutaka Endo, Takashi Kakue, and Tomoyoshi Ito
Opt. Lett. 44(12) 3038-3041 (2019)

Residual D2NN: training diffractive deep neural networks via learnable light shortcuts

Hongkun Dou, Yue Deng, Tao Yan, Huaqiang Wu, Xing Lin, and Qionghai Dai
Opt. Lett. 45(10) 2688-2691 (2020)

Volumetric 3D display in real space using a diffractive lens, fast projector, and polychromatic light source

Christopher Blackwell, Chi Can, Javid Khan, Xianzhong Chen, and Ian Underwood
Opt. Lett. 44(19) 4901-4904 (2019)

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

Supplementary Material (1)

NameDescription
» Visualization 1       The experimental results for the remaining thirty-two locations are shown in Visualization 1, where the maximum PV value of the error after correction is 0.1644 mm, indicating satisfactory improvement is achieved by the neural network.

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

Figures (5)

You do not have subscription access to this journal. Figure files 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

Tables (2)

You do not have subscription access to this journal. Article tables 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

Equations (4)

You do not have subscription access to this journal. Equations 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

Metrics

You do not have subscription access to this journal. Article level metrics 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