Extensive Support System and Knowledge Base for Improvised Geoponics Using Convolutional Neural Network Based Deep Learning
Farming can either be envisaged as a complex multistep process or a linear and recursive feedback loop. In both cases, it requires great effort and cost to produce crops initially. Farmers have extensive knowledge about the crop they are growing, and they must know about the product, but sometimes it so happens that the farmers do not have at par and updated knowledge about the current growth form of the product. The loss per hectare of cultivation has been exponentially increasing over time due to extensive crop damages and failures. Our model suggests an overall algorithm to provide wholesome support and information to the producer may that be a farmer or someone who wishes to take up farming as a hobby. The CNN based model identifies the plant using fine-grain details like location data, timestamp, and state of harvest. Alongside the identification model, a similar model runs parallelly to identify any diseases that might be visible and also provides necessary steps to get rid of the same. Hence providing a safe and low-risk path towards mass producing a certain crop.