Crop classification using low altitude remote sensing platforms

Abstract

Remote sensing has been found to be a valuable tool in evaluating, monitoring and management of land, water and crop resources. Satellite imagery and aerial imagery have wide applications in the field of agriculture, monitoring snow cover trends, wild fire trends, water level rising and forestry. In agriculture, crop health monitoring, yield estimation, classification of crops based on land cover and monitoring of droughts are some common applications of remote sensing. Crop classification in early phenological stages has been a difficult task due to the spectrum similarity of different crops. For this purpose, low altitude platforms such as drones have great potential to provide high-resolution optical imagery where Machine/Deep Learning techniques are applied to classify different types of crops. In this research work, crop classification is performed at different phenological stages. using optical images which are obtained from drone. For this purpose, gray level co-occurrence matrix (GLCM) based features are extracted from underlying grayscale images collected by the drone. In order to classify the different types of crops (wheat, maize, rice, soybean, and winter wheat), Random Forest, Naive Bayes, Neural Network, Support Vector Machine, Convolu tional Neural Network, and Long Short Term Memory Network are applied on the grayscale images and extracted GLCM features. The results showed that the Random Forest algorithm performed much better on GLCM features with an accuracy of 90.91% as compared to grayscale images with a margin of 13.65% in overall accuracy