Non destructive techniques of automatic shape classification of vegetal products

Over the last few years, there has been more and more interest upon carrying out some automatic machines meant to classify both industrial and vegetal products. Shape color and size represent the most important parameters for vegetal products to be assessed via machines vision techniques as a part of automatic processes. Taking into account the large amount of parameters for vegetal products to be reached as a goal of automatic classification, as well as the high degree of fluency of certain parameters, the usual algorithms can easily turn ineffective. Thus, a prototype of an automatic machine hinting to classify vegetal products was projected and brought to life so as to develop and test a series of decision algorithms belonging to artificial intelligence. To classify the shape and size has been developed a back-propagation feed-forward artificial neural network. Part of the result and conclusions are presented through this paper. Key-Words: Machine vision, shape, classification, image processing, neural network