Improving Particle Segmentation from Process Images with Wiener Filtering
While there is growing interest in in-line measurements of paper making processes, the factory environment often restricts the acquisition of images. The in-line imaging of pulp suspension is often difficult due to constraints to camera and light positioning, resulting in images with uneven illumination and motion blur. This article presents an algorithm for segmenting fibers from suspension images and studies the performance of Wiener filtering in improving the sub-optimal images. Methods are presented for estimating the point spread function and noise-to-signal ratio for constructing the Wiener filter. It is shown that increasing the sharpness of the image improves the performance of the presented segmentation method.
Keywordspulp suspension fiber segmentation Wiener filtering machine vision image processing and analysis
- 1.Allied Vision Technologies. Guppy Techical Manual (2009)Google Scholar
- 5.Gonzales, R.C., Woods, R.E.: Digital Image Processing, 3rd edn. Pearson Prentice Hall, London (2008)Google Scholar
- 8.Szeliski, R., Joshi, N., Kriegman, D.J.: Psf estimation using sharp edge prediction. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR (2008)Google Scholar
- 10.Robertson, G., Olson, J., Allen, P., Chan, B., Seth, R.: Measurement of fiber length, coarseness, and shape with the fiber quality analyzer. Tappi Journal 82, 93–98 (1999)Google Scholar
- 12.Sitholé, B., Filion, D.: Assessment of methods for the measurement of macrostickies in recycled pulps. Progress in Paper Recycling 17 (2008)Google Scholar