Machine vision in measurement and control of mineral concentration process

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
Doctoral thesis (article-based)
Checking the digitized thesis and permission for publishing
Instructions for the author
Degree programme
Verkkokirja (13374 KB, 111 s.)
This thesis considers machine vision in the context of the mining, mineral and metal industry (MMMI). Even though MMMI might be seen as a rather conservative industry branch, in many cases it is not. One motivation for constant research and development is the large amount of ore processed on a yearly basis, which means that even a slight improvement in performance can lead to substantial economical benefits. Another point, related more closely to the thesis, is that the development in camera and information technology has enabled the integration of machine vision based applications into many different industry branches, MMMI being one of them. Machine vision and its utilization in measurement and control of a modern flotation plant is studied in detail. The research was started in the late 90's with the development of an image analysis platform for flotation froths, which was later extended to cover multiple flotation cells. The resulting image analysis based variables were studied and new results regarding their usefulness both in single and multi-camera settings were obtained. The most important variables are shown to the plant operators and used in closed loop control. Furthermore, an image history database and a tool for its utilization were created, as well as a new type of froth level measurement technique introduced. The research done with the image analysis of flotation froths provided strong evidence of the importance of the froth colour as an indicator of grade. This motivated further studies carried out with a spectrophotometer, which is a more accurate instrument for colour measurements. As a result, a new type of on-line measurement technique was created to be used as a supplement to existing X-Ray fluorescence (XRF) analyzers to reduce their typical sampling interval of 10-20 minutes to a virtually continuous measurement. Another field of research presented is the particle size distribution analysis of crushed ore from a moving conveyor belt in a contact-free manner, for which two new measurement techniques are presented. This information, when measured already in the mine, can be used in the flotation plant to gain better grinding results, and geologists can use it in mine planning.
machine vision, mine, mining, flotation, control
Other note
  • [Publication 1]: Jani Kaartinen and Heikki Koivo. 2002. Machine vision based measurement and control of zinc flotation circuit. Studies in Informatics and Control, volume 11, number 1, pages 97-105. © 2002 by authors and © 2002 National Institute for Research and Development in Informatics. By permission.
  • [Publication 2]: J. Kaartinen and H. Hyötyniemi. 2003. Determination of ore size distribution with image analysis. In: M. H. Hamza (editor). Proceedings of the IASTED International Conference on Intelligent Systems and Control (ISC 2003). Salzburg, Austria. 25-27 June 2003, pages 406-411. © 2003 International Association of Science and Technology for Development (IASTED). By permission.
  • [Publication 3]: J. Kaartinen and H. Hyötyniemi. 2005. Combining multi-camera-data of flotation circuit with PCA and PLS. In: Graeme J. Jameson (editor). Proceedings of the Centenary of Flotation Symposium. Brisbane, Australia. 6-9 June 2005, pages 121-125. © 2005 by authors and © 2005 Australasian Institute of Mining and Metallurgy (AusIMM). By permission.
  • [Publication 4]: Jani Kaartinen, Jari Hätönen, Martti Larinkari, Heikki Hyötyniemi, and Jorma Miettunen. 2005. Image analysis based control of copper flotation. In: P. Horacek, M. Simandl, and P. Zitek (editors). Preprints of the 16th IFAC World Congress (IFAC 2005). Prague, Czech Republic. 4-8 July 2005. © 2005 International Federation of Automatic Control (IFAC). By permission.
  • [Publication 5]: J. Kaartinen, J. Hätönen, H. Hyötyniemi, and J. Miettunen. 2006. Machine-vision-based control of zinc flotation – A case study. Control Engineering Practice, volume 14, number 12, pages 1455-1466. © 2006 Elsevier Science. By permission.
  • [Publication 6]: Jani Kaartinen, Olli Haavisto, and Heikki Hyötyniemi. 2006. On-line colour measurement of flotation froth. In: V. L. Syrmos (editor). Proceedings of the 9th IASTED International Conference on Intelligent Systems and Control (ISC 2006). Honolulu, Hawaii, USA. 14-16 August 2006, pages 164-169. © 2006 International Association of Science and Technology for Development (IASTED). By permission.
  • [Publication 7]: J. Kaartinen and A. Tolonen. 2008. Utilizing 3D height measurement in particle size analysis. In: Myung Jin Chung, Pradeep Misra, and Hyungbo Shim (editors). Proceedings of the 17th IFAC World Congress (IFAC 2008). Seoul, Korea. 6-11 July 2008. © 2008 International Federation of Automatic Control (IFAC). By permission.
  • [Publication 8]: Olli Haavisto, Jani Kaartinen, and Heikki Hyötyniemi. 2008. Optical spectrum based measurement of flotation slurry contents. International Journal of Mineral Processing, volume 88, numbers 3-4, pages 80-88. © 2008 Elsevier Science. By permission.
  • [Errata file]: Errata of publication 8