Published Paper Details:

India has a big agriculture industry. It is essential for the survival and growth of the Indian economy. India is a major producer of numerous agricultural products. Soil is important for the cultivation of crops. Soil is a dynamic, non-renewable natural resource that is necessary for life. Young Indian farmers frequently struggle with choosing the proper crop depending on the requirements of the soil. As a result, they see a huge drop in productivity. Farmers with practical experience used to cultivate crops in the past. Farmers can no longer choose the best crop based on the characteristics and attributes of the soil. In order to suggest the crop that can be harvested in that particular soil, a recommendation system that makes use of a machine learning algorithm has been created. The user-supplied image of the soil is processed in the proposed system and divided into one of four soil types: clay, alluvial, red, and black. This is achieved using a MobileNetV2 Architecture model. When the soil type is predicted, a number of crops that can be produced there are advised. Our proposed approach maps the soil and crop data to predict the list of crops that will grow well in a given soil. The farmers will therefore find our suggested approach helpful in guiding them to select the best crops for their soil and in educating Our proposed system achieved Training accuracy of 97.34% and validation accuracy of 99.21%.