Abstract For efficient resource management and to guarantee food security, crop yield estimates must be accurate. Deep learning techniques combined with satellite imagery have become a potent method for predicting crop yields in recent years. Deep learning algorithms can extract Data from satellites to provide spatial and temporal information that can be used to analyze crop development patterns and environmental factors. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are two examples of these methods. For a precise estimate of crop production, satellite photography offers useful information on soil characteristics, meteorological conditions, and vegetation indices. The application of deep learning with satellite imagery for crop yield estimation is discussed in general terms in this study, including data collection, pre-processing, model selection, feature extraction, yield prediction, and model validation. The recommended method creates a comprehensive agricultural yield prediction system that links raw data to projected crop yields by fusing deep learning and data mining approaches. Incorporating the Tweak Chick Swarm Optimization method for data pre-processing, the proposed model combines the Visual Geometry Group (VGG) Net classification algorithm with a discrete deep belief network. The model outperforms other models by accurately capturing the baseline data distribution, resulting in an accuracy rate of 97% for predictions.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|