IDENTIFICATION OF SEEDS BY COLOR IMAGING - COMPARISON OF DISCRIMINANT-ANALYSIS AND ARTIFICIAL NEURAL-NETWORK

This study describes the use of colour image analysis to identify four seed varieties. A wide range of kernel measurements was obtained from digitised colour images of whole seed samples of rumex, wild oat, luc erne and vetch. The combination size, shape (including kernel seven in variant moments) and texture parameters is the major element in this i nvestigation. Two pattern recognition approaches were attempted in the classification: stepwise discriminant analysis, which is part of stat istical pattern recognition techniques, and artificial neural network. The artificial neural network was found to outperform discriminant an alysis. With only three inputs, a simple three-layer perception networ k exhibited performances exceeding 99% both in learning and test sets. It is shown that a mixture of features improved classification from 9 2% for size and shape parameters to 99% for size, shape and texture pa rameters. Two species, totally overlapped in the morphometrical space, were well separated by texture. The best characteristics are extracte d from the red channel images. Limitations of neural computing concept s are discussed with respect to seed classification.