Multi-Class Plant Leaf Disease Detection Using a Deep Convolutional Neural Network

Traditional machine learning methods of plant leaf disease detection lack successful performances due to poor feature representation and correlation. This paper presents a novel methodology for automatic using cascaded deep convolutional neural network (CDCNN) which focusses on increasing the correlation factors. It provides distinctive features that gives low intra-class variability higher inter-class variability. CDCNN were performed plant-village database consists 13 classes tomato, potato, pepper bell diseases; DCNN model performs better with an overall accuracy, recall, precision 98.50%, 0.98, 0.97 respectively. Additionally, performance proposed algorithm is evaluated real time cotton bacterial blight, miner, spider mite diseases 99.00% accuracy. The outperforms well compared traditional models able detect present in leaves plant.