CCNN-PDC: Design and Build a Customized CNN Classifier Model for Plant Disease Classification and Detection Using Deep learning

Agriculture is considered as the most essential & important part as the income source of the farmers around the world, however they do face the annual production loss because of the plant disease. The tailed twist here is that those loss which farmers face can be actually reduced by the detection of those plant diseases at early. Manually doing the recognition can be whole tiresome, also the recognition done won't be always accurate as the two possible disease early blight & late blight are similar in state. To unmute such problems, we have devised our own customized CNN model also various strategies of Deep Learning (DL) & Convolution Neural Network (CNN) Sky-rocketed in recent times as they are helpful in accurate diagnosis in less amount of time & even combining these methods with Image processing, the images of different plant with its respective diseases are used to identify the plant diseases. Dataset for the current research was besotted from the Kaggle. Google cloud Functions is used for deployment and Fast API is used for testing purpose of our model. The web app interface which we had developed can take the input as an image & it will conclude whether the input provided of the plant leaf is healthy or not if it turns out to be the known disease it will show the plant with its respective disease & will provide the solution to set your plant disease free. We assessed the performance of our own Customized CNN model for plant disease detection against the popular VGG-16 architecture in order to achieve our primary goal.