Generic Paddy Plant Disease Detector (GP2D2)

Rice is the primary food for almost half of world’s population, especially people Asian countries. There a demand to improve quality and increase quantity rice production meet requirements increasing population. Bulk cultivation crops need appropriate technology assistance over manual traditional methods. In this work, six popular Deep-CNN architectures, namely AlexNet, VGG-19, VGG-16, InceptionV3, MobileNet, ResNet-50, are exploited identify diseases in paddy plants since they outperform most image classification applications. These CNN models trained tested with Plant Village dataset classifying plant images into one four classes namely, Healthy, Brown Spot, Hispa, or Leaf Blast, based on disease condition. The performance chosen architectures compared different hyper parameter settings. AlexNet outperformed other convolutional neural networks (CNNs) multiclass task, achieving an accuracy 89.4% at expense substantial number network parameters, indicating large model size AlexNet. For developing mobile applications, ResNet-50 architecture was adopted CNNs, it has comparatively smaller parameters comparable 86.1%. A fine-tuned supported app, “Generic Paddy Disease Detector (GP2D2)” been developed identification commonly occurring plants. This tool will be more helpful new generation farmers bulk productivity paddy. work give insight detection task can extended too.