Abstract

The applicability of remote sensing is improving hand in hand with time. Various research works focus on remote sensing technology, as it is one of the hottest research topics. This paper is all about satellite image crop classification. The crops being present in a particular location is differentiated by means of a classification algorithm. However, it is difficult to attain reasonable accuracy rates, as the images are captured from a greater altitude. This research article focuses to present a satellite image classification system for distinguishing between the crops being present in the agricultural area. To achieve the research goal, the entire work is broken down into satellite image pre-processing, feature extraction and classification. The satellite images are mostly affected by noise and poor contrast. These issues are addressed by employing bilateral filter and adaptive histogram equalization technique. The Gabor Local Vector Pattern (GLVP) based Scale Invariant Feature Transform (SIFT) features are extracted from the pre-processed images. The crops being present in a location are distinguished by means of ensemble classifier, which is a combination of k-Nearest Neighbour (k-NN), Support Vector Machine (SVM) and Extreme Learning Machine (ELM). The performance of the ensemble classifier is compared with the individual classifiers, and the ensemble classifier outperforms the other classifiers in terms of classification accuracy, sensitivity and specificity rates.