AgriSynth: Machine Learning - Driven Crop Intelligence, Fertilizer Rationalization and Disease Diagnostics

Agriculture is important to feeding the world's growing population, especially in the face of climate change. Machine learning technology has the potential to greatly improve agriculture by helping farmers make more educated decisions. However, current systems have considerable limitations. Many fertilizer recommendations only address the key nutrients NPK: nitrogen, phosphorus, and potassium, fail to account for water availability, and rely on out-of-date historical data. This study would close these gaps by producing a machine learning system that can help farmers with crop recommendations, fertiliser guidance, and plant disease detection. The algorithm considers micronutrients and organic matter in addition to NPK when recommending fertiliser. It also incorporates water availability and real-time meteorological data to provide more precise crop suggestions, especially important for drought-prone areas. Deep learning methods analyze plant photos to detect diseases early on. Our findings indicate that by incorporating micro nutrients, water data, and real-time climate updates, the system can deliver more precise crop and fertilizer recommendations. The Random Forest and XGBoost models outperform, and the disease detection component helps to prevent crop losses by intervening early. The key takeaway from this study is that merging machine learning with large amounts of data can significantly improve farming techniques, raise crop yields, and provide farmers with reliable insights, thereby enhancing food security and sustainable agriculture.