Deep Convolution Network Analysis for Crop Growth Prediction

Farmers have been facing a lot of difficulties in yielding crops in a healthy state. There has been work where Remote sensing technologies employed for getting real time information towards crop growth. Also work has been done in employing Machine and Deep learning for predicting the stages of apple growth, plant phenological and olive tree phenological stages. But in none of work, there has been an attempt to predict the stage of crop growth with a resource optimised Deep Convolution Neural Network (DCNN) model. So we in this work have performed comparitive analysis of three CNN models like VGG19, Xception and Mobile NetV2 for rice crop growth as case study in terms of accuracy, error and memory usage for resource constrained devices. This could help the farmer in assessing the stages of crop growth to take early action before it drastically affects the entire farm