Desenvolvimento e avaliação de um protótipo classificador de tábuas usando técnicas de visão artificial

Abstract

Lumber quality is determined by the defects presented and other characteristics such as: size, position, amount and type. The purpose of this research was to develop and evaluate a prototype to classify eucalyptus lumber using digital images. This prototype was built with a conveyor belt where the lumbers are inserted for image acquisition. Either the Brazilian standard (ABNT) or the commercial rule can be used for classification. The process can be followed in a microcomputer that shows the lumber image with its final grade. The overall accuracy rate in the classification process was 64.3% using the ABNT norm, and 81.0% percent using the commercial norm. Productivity of the developed prototype was 7.9 m 3 h -1.