Abstract
A structured-light system calibration method based on a neural network was proposed. By using the method of projective transformation and error compensation, the mapped relation between the camera image plane and the projector image plane was obtained. Then, with the relation and the camera image-coordinates, the corresponding projector image coordinates were calculated. So, a three-layer neural network was constructed. For this network, the inputs are two image coordinates and outputs are 3D world coordinates. The training set consists of two image coordinates and 3D world coordinates of calibration points. Then, the neural network was trained by Back Propagation (BP) algorithm while the system model was fitting with it. When the process of the training was finished, the calibration was also accomplished. The results of the experiments prove that the method proposed in the paper reveals a higher degree of accuracy comparing with the conventional methods, and reduces the complexity of the model and simplifies the process of calibration. Besides, it can be applied in various conditions gererally.
Original language | English |
---|---|
Article number | 0512002 |
Journal | Guangzi Xuebao/Acta Photonica Sinica |
Volume | 45 |
Issue number | 5 |
DOIs | |
State | Published - 1 May 2016 |
Externally published | Yes |
Keywords
- Backpropagation
- Cameras calibration
- Model
- Neural network
- Projective transformation
- Structured-light
- Surface measurement