A real time and completely automated colour classification tool for woollen cloths to be recycled is proposed. The tool uses the combination of a statistical method, called Matrix Approach, and an Artificial Neural Network (ANN) based method, to correctly classify the cloths by respecting the selection criteria provided by human know-how. The error in classification is within 3 %, i.e. less than the error resulting from the use of an expert human operator. The comparison between the results obtained by the colour classification tool and the human expert classification is also presented.
A color classification tool based on neural networks / R. Furferi; L. Governi;. - ELETTRONICO. - (2003), pp. 0-0. (Intervento presentato al convegno XIII ADM - XV INGEGRAF International Conference on Tools and Methods evolution in engineering design tenutosi a Napoli nel 3-6 Giugno 2003).
A color classification tool based on neural networks
FURFERI, ROCCO;GOVERNI, LAPO
2003
Abstract
A real time and completely automated colour classification tool for woollen cloths to be recycled is proposed. The tool uses the combination of a statistical method, called Matrix Approach, and an Artificial Neural Network (ANN) based method, to correctly classify the cloths by respecting the selection criteria provided by human know-how. The error in classification is within 3 %, i.e. less than the error resulting from the use of an expert human operator. The comparison between the results obtained by the colour classification tool and the human expert classification is also presented.
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
