Plant Disease Identification Tracking and Forecasting Using Machine Learning
M. Sowmya Vani, S. Girinath, V. Hemasree, Lomada Hars Havardhan, Poli Sandhya
Abstract
How safe our food supply is and how productive our agriculture is are both significantly impacted by plant diseases. For prompt cures and efficient management of plant diseases, accurate diagnosis, close observation, and foresight are necessary. Methods like deep learning, feature extraction, and picture recognition are frequently used in disease detection. For farmers, agronomists, and decision-makers, machine learning can be used to distinguish, monitor, and forecast illness affecting plants. The Plant Village Data set is meticulously segmented for disease prediction training and testing; as a result, several plant species are acknowledged and given new names to make an accurate database. The classifier is then trained using training data, and the output will then be detected with the highest accuracy possible. Therefore, this study presented a CNN-based system for disease detection in plants and also evaluated the overall performance of various classifiers at the study's records set to decide which had the most accuracy. The performance and usability of machine learning models must be improved through ongoing research and development if we are to eventually see more effective and sustainable farming practices. The created model also maps the soil and crop database and recommends appropriate crops depending on the number of nutrients available in the soil, enabling farmers to choose the right crops to be sown in their fields. Therefore, it is important to identify crop diseases as soon as possible. Farmers will profit from using a quick, creative approach and a crop recommendation system.