Optimizing Crop Spraying with Smart Drones using Deep Neural Networks and Image Processing

Due to climate change, environmental issues, and the necessity to nourish a growing human population, intelligent drones for agricultural spraying are gaining popularity. The utilisation of IoT smart drones in agriculture enhances food security by facilitating more precise operations, optimising resource utilisation, and reducing environmental impact. The proposed process consists of three steps: data preparation, feature extraction, and model training. Agricultural images are subjected to scaling, denoising, augmentation, and background elimination in the preprocessing phase. Significant attributes such as hue, shape, density edge, and HOC are identified by the feature extraction module. The study employs an ABiLSTM model that utilises historical data to determine optimal timings for crop spraying, hence enhancing prediction accuracy. The results indicate that the ABiLSTM model effectively addresses challenges in complex feature classification and prediction, achieving an impressive 95.62% accuracy rate. This study illustrates how AI-powered smart drones may transform precision agriculture, offering a sustainable and economical solution to the issues encountered by modern farmers. This research establishes a foundation for future AI-driven agricultural techniques that can enhance crop management through deep learning and the Internet of Things. This will assist farmers in fulfilling future food supply demands.