Cullets (Glass Fragments) Quality Control by Artificial Vision: a Textural Based Approach

Glass fragments (cullets) to be recycled present different market values according to their colour. Glass recycling plants perform cullets sorting mainly discriminating coloured glasses from white and half white glasses. Cullets which are collected without distinctions of colour, can be used primarily for the production of green glass and only in part for the production of yellow glass. The production of white glass requires that only cullets of that colour be employed. At present, machines for the separation of cullets according to colour are not capable of producing an efficient classification of all the different types. In this paper are analysed the possibility that could be offered by the adoption of a colour imaging based approach to realise cullets sorting, analysing the textural attributes. This study was addressed on the effects that cullets surface status and characteristics produce on the detected colour-textural characteristics and they can influence the further classification.