CNN Based Detection of Healthy and Unhealthy Wheat Crop

The diagnosis of leaf diseases depict a wide range of information about the crop status. Diagnosis plays a very crucial ro le on the amount of resources available the farmt It gets prioritize because it effects the nation's gross domestic product (GDP) directly. It is beneficial for the analysis of crop in early stage to Verify the efficient crop yield. Automatic disease identification is one of the interesting and challenging problems in computer vision due to its potential applications. In this paper, a novel method has been proposed for disease identification. The proposed method suggests a feature extraction solution for the identification of healthy and unhealthy wheat plant. A deep convolutional neural network (DCNN) and transfer learning approach is used to train the model. Different CNN models like VGG16, VGG19, AlexNet, ResNet-34, Resnet-101, ResNet-50 and ResNet-18 are used to train our model for obtain ing better results. Accuracy achieved while training the model is up to 98%. Results have showed that the technique used in this paper is beneficial to farmers so that they can identify the spoiled area of a crop and utilizes the resources to increase their productivity.