Recognition of seed varieties using neural networks analysis of electrophoretic images

The paper presents a method for seed variety recognition using one-dimensional electrophoresis gels. It employs the Time-Delay Neural Network (TDNN) and the Temporal Organisation Map (TOM) which are initially developed for speech recognition. These neural networks can be trained to recognise the presence of a phoneme or a word in speech by reference to the sound pattern over a sequence of time steps. Electrophoresis creates a set of bands in the gel, caused by migration of protein from the seed. Each seed variety generates a characteristic pattern. The bands are made visible by staining. They can then be imaged and digitised to create an input to a TDNN or a TOM, which treats the variation with distance along the lane in the same way as the time sequence for which it was originally employed. In that way the characteristic signature of a seed variety can be recognised. Furthermore, a set of images, each containing 10 to 15 lanes, was used to train and test the performance of neural networks in recognising cereal varieties. The networks could achieve a recognition rate of 98 per cent, provided that the gel was not distorted or cracked during heating or drying.