Development And Simulation Of Ios App To Detect Cotton Plant Diseases Using Machine Learning Techniques

Cotton is the most commercial crop in India. The farmers can get good capital when the production of cotton is good. The cotton crop is easily exposed to the disease which reduces productivity. In this paper, an iOS app is developed which will detect the cotton disease by taking a photo of the disease leaf. The app can work without the internet and the app will show the crop disease details. The app also recommends the remedy of the disease. The farmers can see the disease history in the app. The app supports the two types of image filters to handle different kinds of images in a different environments. The app detects the fungal leaf spot and boll rot disease. The app is based on the iOS ML model which is used for prediction. The ML model is created by using the CreateML mac application. All the labelled data sets are added to the CreateML application for image classification. The mac app will then train the model and evaluate the model accuracy. The accuracy of the model is 91%. The native core ml Apple framework is used in app development. The app is based on the VIPER iOS app architecture which has different layers with their responsibility. Further development can be done to detect more diseases and server-based implementation can also be done. The CoreML apple framework is used for ML techniques to identify the disease. The app is developed using the Swift