The monitoring of the plant diseases has advanced dramatically and is taking us in a path towards real time accurate and scalable diagnostics due to the advances in the technology. To examine the integration of Machine Learning (ML), remote sensing and Mobile Applications for improved Real Time plant disease monitoring, this research paper has been written. The limitation of traditional methods of plant disease detection include the factors of inaccuracy, slow response time, and slow scalability that made it challenging for timely intervention and large agricultural losses. This paper tackles these challenges by dealing with advanced plant disease monitoring and management procedures. The basic datasets are plant images and environmental data, and they have already shown the ability to provide useful information when analyzed with ML Algorithms such as CNN and some ensemble methods. CNNs can determine causes of disease so subtle that they are invisible to traditional modes of detection. The ensemble methods improve the accuracy and robustness of the diagnostic through using multiple ML models. Satellite imagery and drone based sensors have impact on plant health, and environment condition. Remote sensing data, when combined with ML models, allow us to continue monitoring the crop conditions continuously, and unlike with any kind of sensor, early warnings of disease outbreaks or of excess resource usage are provided. Mobile applications with smartphone cameras and ML algorithms that give on the go disease diagnoses help the farmers make instant actionable information for early intervention. Recent advancements and case studies introducing the use of these technologies for real time plant disease monitoring are found in this paper. Most importantly, it suggests that using the new technologies to the maximum, enables us to develop accurate and scalable disease management. ML, remote sensing and mobile technologies have a high integration potential for disease plant monitoring, improving agricultural efficiency and food security.
Mishra et al. (Fri,) studied this question.