Automatic identification of flower diseases using artificial neural networks

The floral industry has increasingly become one of the most important sectors for export earnings, especially in developing countries. However, during the cultivation process there may be a number of challenges that affect it, one of which is flower disease. This paper presents an automatic identification of of flower dieases based on image processing techniques. In view of this, normal and diseased flower images are acquired to create a knowledge base where images are pre-processed and segmented to identify the region of interest. Texture features of images are extracted using Gabor feature extraction, from which we computed seven different measures of dispersion and central tendency with the purpose of reducing the dimensionality of features. Then, an artificial neural network is trained with seven input features extracted from individual images and eight output nodes representing eight classes of diseases considered in this work. Unkown samples of flower images are then tested based on the training model and we achieved an average accuracy of 83.3% in the identification of the flower diseases.