Desarrollo de un sistema de reconocimiento para piezas faltantes en moldes en la industria de alimentos mediante el procesamiento de imágenes con matlab

Abstract

The chocolate industry transforms cocoa production into finished products that are then exported. During production there is a risk of moulds breaking as mechanical parts of the equipment that cause moulds to deteriorate are involved in the manufacturing process, the probability of a piece of mould falling into a tablet and reaching a final consumer is high, which can generate legal penalties and economic losses for the company by creating a risk for the consumer, to reduce this, the present project makes an analysis of the variables involved in the chocolate injection process in the line super cavemil 800 and implements a system of image capture of the molds that make up the production machine and through the Matlab googlenet neural network image processing allows the recognition of missing parts in the molds can thus alert the operator of the machine the presence of an anomaly in them activating an alert protocol for production.