Crop production rate estimation using parallel layer regression with deep belief network

Ensuring food productivity is an important issue for the developing countries like India, where more than one third of the people is live in poverty. Season crop production estimations are vastly identified as an important input for investigating food balance sheets and production shortfalls. Crop production estimation and assessment is done worldwide on a regional basis to enable high production and cost reduction in crop yield estimation. The aim of this method is to the productivity of the crop, using various factors like soil type, season type, water availability and risk factor. In this scenario, Parallel Layer Regression (PLR) along with Deep Belief Network (DBN) strategy is proposed to perform crop productivity estimation. Here, DBN strategy is generated for top five growing crops in Karnataka namely, rice, ragi, and pulses. The proposed methodology forecasts each area in the applicable database into one of the five crops. Finally, the experimental results show that the method has strong potential for accurate crop productivity prediction in terms of accuracy (ACC), sensitivity (SEN) and specificity (SPE) and also this method performance has verified in real time data and people interactions.