Deep Network with Score Level Fusion and Inference-Based Transfer Learning to Recognize Leaf Blight and Fruit Rot Diseases of Eggplant

Eggplant is a popular vegetable crop. yields can be affected by various diseases. Automatic detection and recognition of diseases an important step toward improving crop yields. In this paper, we used two-stream deep fusion architecture, employing CNN-SVM CNN-Softmax pipelines, along with inference model to infer the disease classes. A dataset 2284 images was sourced from primary (using consumer RGB camera) secondary sources (the internet). The contained nine eggplant Experimental results show that proposed method achieved better accuracy lower false-positive compared other learning methods (such as VGG16, Inception V3, VGG 19, MobileNet, NasNetMobile, ResNet50).