For many individuals, one of the most significant sources of income is agriculture. The COVID-19 epidemic has also had a severe impact on conventional farming. The FAO reports that some people's access to food is being restricted by border closures, quarantines, and interruptions in supply chains, particularly in nations that have been severely affected by the virus or that already have high levels of food poverty. In 2016, the market for indoor farming technology was estimated at $23.75 billion, and by 2022, it is expected to reach $40.25 billion. When compared to traditional agricultural techniques, yields are often substantially greater. Instead of growing in only two dimensions, crops planted indoors may do so all year round, regardless of the weather outside. A key study area in smart horticulture is the intelligent diagnosis and categorization of greenhouse plant diseases. However, because plant diseases frequently arise organically, they have an impact on a lot of farmers. If necessary precautions are not followed, illnesses can be dangerous to plants and affect the productivity, quality, or quantity of the final output. Because of this, plant disease detection and prevention are major issues that should be taken into account in order to boost output. A better plant disease-recognition algorithm is required to correctly identify plant diseases under challenging environmental settings. An innovative hybrid classification system is utilised in this work to identify the tomato disease. The hybrid FOA-SVM classification method is based on Fruitfly Optimization Algorithm (FOA) & Support vector machine (SVM). The model is utilised on tomato plants for precise disease identification and improved diagnosis. For all types of plants, Support Vector Machine is shown to be the most effective classifier. The experimental findings demonstrate that our hybrid technique is more accurate than other methods already in use and has a substantial impact on the categorization of illnesses that are similar.

Feature Extraction, Segmentation, Classification, SVM, FOA