Published Paper Details:

Learning software for machines is quickly becoming the most popular and widespread form of agricultural technology. Artificial machine learning is one of the sectors of the agriculture industry that is growing at the fastest rate. The application of artificial technologies in the field of agriculture has led to improvements in accuracy as well as the finding of answers to previously unresolved issues. In agriculture, the application of machine learning should work towards improving crop quality while also increasing plant output. Cotton is the most essential crop for producing fiber, not only in India but also around the world. The cotton textile industry stands to benefit from the acquisition of this since it is the raw material used in the production of fiber (cotton fiber). Infections caused by bacteria, fungus, and worms, as well as unregulated agricultural practices that are harmful to the leaves, are the primary challenges that cotton growers confront. Therefore, in order to prevent a low yield, farmers need to be aware of the crop diseases that have hurt the crop in the past. The consistent use of pesticides, together with the evolution of pests into organisms that are increasingly resistant to the effects of chemical control, has resulted in yields that have remained flat or fallen, in addition to a decline in the quality and productivity of the soil. Although a great number of cutting-edge machine learning models have been developed and examined, widespread application of these models in agricultural contexts has not yet been achieved. Transfer learning is a method that can be used to adapt and apply existing models in order to improve disease detection in plants and crops. In order to diagnose diseases that affect cotton plants, we have utilized contemporary methods and presented a hybrid methodology. In a hybrid method, contemporary computer vision models such as Faster RCNN and Single Shot Detector (SSD) are employed to recognise plant leaves and locate ROI (Region of Interest).