A survey on Deep Learning Architectures for effective Crop Data Analytics

Deep learning has emerged as a precise tool for image and non-image-based data analytics in various domains. Smart farming is a major domain where deep learning approaches have been effectively applied for crop data analytics. Crop monitoring is a significant task in smart farming to enhance agricultural productivity. Accurate prediction of crop yield with limited availability of information related to environmental parameters is the main challenge. Crop diseases are often the major reason for yield loss. In developing countries, rural farmers are not equipped with modern techniques for real-time crop disease detection and thus fail to implement corrective actions. Recently, deep learning approaches have proven to be more effective in generating more accurate results in crop data analytics. In this study, research works pertaining to deep learning-based models applied in crop yield estimation and crop disease detection are reviewed.