Development of a custom machine vision camera for froth flotation

Date
1999-06
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Flotation is one of the most important and versatile mineral processing techniques in the mining industry. Unfortunately, Froth Flotation processes are subjected to a wide variety of process disturbances; some of which are caused by changes in mineral characteristics, and others by variation in operating procedures. The aim of this thesis is to develop a custom Machine Vision Camera for Froth Flotation plants. Recent research shows that Machine Vision techniques can alleviate many of the difficulties faced by conventional flotation control procedures. The goal of the Machine Vision Camera is to assist in the further development of optimal control strategies for Froth Flotation plants based on the visual features of froth. As additional assistance in this ongoing process, Discrete Image Cross-Correlation is proposed as a tool to visually extract the two most important parameters of froth. The use of dedicated custom hardware based on FPGA and Transputer technology to implement the Discrete Image Cross-Correlation function for a batch of flotation cells is also presented. Final results showed that a market-standard PC will prove simpler, faster, and more cost-effective than similarly priced dedicated custom hardware.
AFRIKAANSE OPSOMMING: Flotasie is een van die belangrikste en veelsydigste mineraal prosessering tegnieke in die myn bedryf. Ongelukkig word skuim flotasie prosesse aan 'n verskeidenheid proses versteurings onderwerp waar sommige deur veranderings in mineraal eienskappe en ander deur variasies in die bedryf van die flotasie proses veroorsaak word. Die mikpunt van die tesis is die ontwerp van 'n doelgerigte Masjien Visie Kamera vir skuim flotasie prosesse. Huidige navorsing to on dat Masjien Visie tegnieke baie van die probleme wat met die konvensionele beheer van flotasie ondervind wordkan oorkom. Die doel van die Masjien Visie Kamera is om die verdere ontwikkeling van optimale beheerstrategiee wat op die visuele eienskappe van die flotasie skuim gebaseer is te ondersteun. As addisionele bystand tot hierdie voortdurende poging word Diskrete Beeld Kruiskorrelasie as 'n visuele metode ondersoek om die twee belangriksteeienskappe van flotasie skuim te onttrek. Die gebruik van doelgerigte hardeware gebaseer op Transputer en FPGA tegnologie vir die uitvoering van die Diskrete Beeld Kruiskorrelasie metode vir 'n groep flotasie selle word ook ondersoek. Finale resultate toon dat'n standaard, kommersiele PR makliker, vinniger en meer koste effektief sal wees as doelgerigte hardeware vir dieselfde prys.
Description
Thesis (M.Ing.) -- University of Stellenbosch, 1999.
Keywords
Flotation, Computer vision -- Industrial applications, Cameras, Dissertations -- Electronic engineering
Citation