Analysis on Prediction of Crop Diseases Using TensorFlow with Keras and OpenCV Technique of Deep Learning

Crop disease is a harmful deviation from the normal growth of plant which affects its appearance, function, or productivity and causes great damage in agriculture resulting in significant yield losses. Prediction of crop diseases can be important for agriculture field, as it can help farmers take preventive measures to protect crops and ensure a good yield. There are various methods that can be used to predict crop diseases, such as examining the plant for symptoms, using weather data to identify conditions that are conducive to the development of diseases, and implementing early warning systems that can alert farmers to the presence of potential diseases. By predicting and preventing crop diseases, farmers can improve the health and productivity of their crops, leading to better yields and increased profitability. This research paper proposes a model whose purpose is to classify the type of disease. Deep learning techniques are used in the context of categorizing plant diseases. Dataset of diseased and healthy crops is used. Leaf images of diseased crop are infused as training set in a deep learning model. TensorFlow with keras and OpenCV are used for prediction of crop disease. OpenCV is used to detect the pattern in the crop leaf image and translate it into data on which ML model can be build using keras. Firebase is used to efficiently manage the large number of images. The trained model achieved the accuracy of 97.80%. Model predicts the disease name, possible causes, and the solution.