Detection of Cotton Plant Disease for Fast Monitoring Using Enhanced Deep Learning Technique

Disease detection in agriculture has become very crucial in today’s deteriorating climatic conditions. Cotton production plays a significant part in our country’s economic growth for being a major cotton producer in the world. Monitoring the large cotton fields manually becomes a tiresome activity for farmers. To leverage advancements in technology, deep learning and image processing techniques are being used to detect diseases in crops at the onset before time and thus preventing them from further harm. In this work, we have done analysis of transfer learning techniques for disease detection and proposed a sequential deep convolutional neural network for accurately classifying fresh and diseased plants. The proposed model gives better accuracy and performs much faster than pre-trained models like VGG16, ResNet50, and ResNetl52V2. So, monitoring of large fields can be done using the proposed model for faster diagnosis and treatment for better production.