Road Extraction from High-Resolution Satellite Images using Deep Learning Model

Abstract

Data Extraction has always been one of the major interests of several companies and government agencies for obtaining information through satellite images. The Data extraction in the form of road extraction from satellite images has played a very important part in navigation, military, national security, agriculture, and understanding the geology of the earth's surface. We are introducing ReLU as the classification function, with the soft-max function. In this paper, we present an automated model for road extraction from a high-resolution satellite image. Here, we are implementing Rectified Linear Unit through Segmentation for obtaining roads out of satellite images with pre-processing, training, and validation of input data. And re-define the quality aspect of satellite images.