Machine-vision-based control of zinc flotation — A case study

It is widely accepted that mineral flotation is a very challenging control problem due to chaotic nature of process. This paper introduces a novel approach of combining multi-camera system and expert controllers to improve flotation performance. The system has been installed into the zinc circuit of Pyhäsalmi Mine (Finland). Long-term data analysis in fact shows that the new approach has improved considerably the recovery of the zinc circuit, resulting in a substantial increase in the mill’s annual profit. r 2006 Elsevier Ltd. All rights reserved.

[1]  Jani Kaartinen,et al.  Image analysis based control of zinc flotation - a multi-camera approach , 2002, 7th International Conference on Control, Automation, Robotics and Vision, 2002. ICARCV 2002..

[2]  Jari Hätönen,et al.  Image analysis based control of zinc flotation , 2001 .

[3]  W. Godwin.  Article in Press , 2000 .

[4]  Chris Aldrich,et al.  The interpretation of flotation froth surfaces by using digital image analysis and neural networks , 1995 .

[5]  G. De Jager,et al.  A technique for automatically segmenting images of the surface froth structures that are prevalent in industrial flotation cells , 1992, Proceedings of the 1992 South African Symposium on Communications and Signal Processing.