Abstract
Image measurement based on machine vision is a promising method for the precise measurements of machine parts. As for a large sized workpiece that exceeds the FOV (field of view) of the camera, image mosaic must be performed to implement measurement. The problematic behavior of image mosaic is the contradiction between the precision of mosaic and the complex of algorithm adopted. Mutual information as a similarity metric has been explained and introduced into the mosaic operation in this paper. To obtain a satisfactory mosaic effect and measuring precision, a machine vision based configuration that integrates the illumination setup and motion control schemes was built to ensure the accuracy. The precise motion control system was utilized to limit the search space and speed up the operation. Experimental results show that the precision of the image mosaic with mutual information as a similarity metric meets the requirements.
Publisher
Trans Tech Publications, Ltd.
Reference5 articles.
1. Charsten Steger, Markus Ulrich and Christian Wiedemann: Machine Vision Algorithms and Applications. Tsinghua University Press, Beijing, China. (2008), pp.1-2.
2. Q.B. Zhang, B. Luo, W. S, et al. Registration for feature point sets based on affine transformation [J]. Journal of Image and Graphics. (2003), No. 10, pp.1121-1125.
3. Josien P.W. Pluim*, J.B. Antoine Maintz, et al. Image Registration by Maximization of Combined Mutual Information and Gradient Information. IEEE Translation on Medical Imaging. Vol. 19(Aug. 2000), No. 8, pp.809-814.
4. Anthony, Amankwah, Lofffeld, Otmar. Image registration using a combination of mutual information and spatial information. IEEE International Geoscience and Remote Sensing Symposium, IGARSS, (2006), pp.4012-4016.
5. X.Y. Chen. Image Registration and Implemention on MATLAB. Publish House of Electronics Industry, China. (2009) pp.148-149.