Use of genetic artificial neural networks and spectral imaging for defect detection on cherries

A machine vision system was created to identify different types of tissue c haracteristics on cherries. It consists of an enhanced NIR range vidicon bl ack and white camera (sensing range 400-2000 nm), a monochrometer controlle d light source, and a computer. Multiple spectral images of cherry samples were collected over the 680-1280 nm range at increments of 40 nm. Using the spectral signatures of different tissues on cherry images, artificial neur al networks were applied to pixel-wise classification. An enhanced genetic algorithm was applied to design the topology and evolve the weights for mul ti-layer feed forward artificial neural networks. An average of 73% classif ication accuracy was achieved for correct identification as well as quantif ication of all types of cherry defects. No false positives or false negativ es occurred, errors resulted only from misclassification of defect type or quantification of defect. (C) 2000 Elsevier Science B.V. All rights reserve d.