An approach to cereal grain classification

It is found that classification of cereal grain using morphological features extraction is time consuming and required preprocessing techniques like image segmentation and morphological features extraction. Earlier research has made classification of cereal grains using Back Propagation neural network. Here the classification is carried out easily for different types of cereal grains and also the classification is made among white rice samples whose colour are all the same i. e. white. The colour and texture features are extracted and base on these features the grains are classified using Back Propagation neural network and the result is compared using Hopfield network. From the result it is found that Hopfield network is a better choice for cereal grain classification. This approach may be used for content base cereal grain image retrieval.