Crop and Fertiliser Recommendation System for Sustainable Agricultural Development

Agriculture is the main source of income for the people living in India and vital to our Indian economy. In recent years, agriculture struggles to meet the demand of population of our country which grows rapidly, and various agricultural problems such as pest detection, plant disease detection and classification, soil classification, cultivation of right crops, crop yield prediction, weed detection, environmental prediction, seed quality prediction and classification and animal intrusion detection are the major threats to farmers. An efficient system is required for monitoring agricultural problems and supporting farmers to identify the crops to be cultivated. Soil is the most important for the living of people on our earth which is a root source for agriculture. Farmers are depending on soil for growing crops which is a warehouse of minerals. Soil is of different types in India. Soil properties and the nature of the soil can vary from one location to another location. The same soil properties cannot be found even within a few distances. Each soil can have different levels of minerals, nutrients and organic matter and can have different characteristics based on the location. So, farmers need to know the soil types and features of various kinds of soils to understand which crops are to be cultivated in that particular soil type in different climate conditions and what kind of pesticides and fertilisers can be used for better crop yield. IoT, machine learning and deep learning techniques are emerging techniques and support smart farming. This study analysed articles on soil classification and crop recommendation systems to classify the soil based on the macro- and micronutrients available in the soil and predict the suitable crop to be grown in different types of soil based on soil data.