ODNet: Optimized Deep Convolutional Neural Network for Classification of Solanum Tuberosum Leaves Diseases

Solanum tuberosum (Potato) is a highly cultivated and consumed crop across the globe. The consumption of the potato is increasing with the increase in population. But the yield of the potato is affected by fungi, pathogens, and insects. The fungus affects the crop yield highly as it occurs at every stage of plant growth. The two crucial diseases caused by the fungus are classified as early blight and late blight. Continuous monitoring of the crop and classifying the disease is difficult for farmers. This paper aims to design a model to classify the diseases using potato leaves images which reduces the cost and time to the farmers. An optimized deep convolutional neural network (ODNet) has been proposed to identify potato leaf diseases. The performance of the ODNet is optimized by tuning the hyperparameters of the model using the Slime Mould Optimized (SMO) algorithm. The performance of the optimized model in classifying the disease leaves is significantly increased compared with the non-optimized model. The accuracy of the proposed model is 96.67 % which is better than the state-of-the-art models.