Abstract : An approach referred to as “smart farming” involves high-end technological advances in modern farming by collecting data from various sources such as robotics, sensors, broadcasts on social media, etc.Plant diseases are usually passed through viruses and insects that are pests that may substantially decrease production when they aren’t immediately handled. Additionally to keeping track of soil quality, this study proposes a system to detect and remove diseases on cotton plants. Though there have been many advances in smart farming down to this moment using image processing, data mining, IOT, etc., machine learning has emerged as the area that is developing most rapidly. In real-time, machine learning applications utilizing supervised or unsupervised approaches are currently in more significant popularity. The conventional farming method includes gathering data, managing unpredictable weather, covering diseases in pesticides, and other behaviors that threaten agricultural workers, particularly in drought-stricken areas. Considering the present state of affairs with conventional farming, there has been an essential demand for predictive data in agriculture which could help farmers to recognize their existing issues and implement necessary measures.