Road Extraction from High-Resolution Remote Sensing Imagery Using Refined Deep Residual Convolutional Neural Network

Lin Gao,Weidong Song,Jiguang Dai,Yang Chen

doi:10.3390/rs11050552

Abstract

Road extraction is one of the most significant tasks for modern transportation systems. This task is normally difficult due to complex backgrounds such as rural roads that have heterogeneous appearances with large intraclass and low interclass variations and urban roads that are covered by vehicles, pedestrians and the shadows of surrounding trees or buildings. In this paper, we propose a novel method for extracting roads from optical satellite images using a refined deep residual convolutional neural network (RDRCNN) with a postprocessing stage. RDRCNN consists of a residual connected unit (RCU) and a dilated perception unit (DPU). The RDRCNN structure is symmetric to generate the outputs of the same size. A math morphology and a tensor voting algorithm are used to improve RDRCNN performance during postprocessing. Experiments are conducted on two datasets of high-resolution images to demonstrate the performance of the proposed network architectures, and the results of the proposed architectures are compared with those of other network architectures. The results demonstrate the effective performance of the proposed method for extracting roads from a complex scene.

Highlights

Roads play a key role in the development of transportation systems, including the addition of automatic road navigation, unmanned vehicles, and urban planning, which are important in both industry and daily living [1]
The heterogeneity in remote sensing images restricts many existing algorithms that depend on a set of predefined features extracted using tunable parameters. (ii) The objects in images are blocked by obstacles, either through shadow occlusion or visual occlusion
The method consists of two major stages: refined deep residual convolutional neural network (RDRCNN) and postprocessing with a tensor voting (TV) algorithm

Summary

Introduction

Roads play a key role in the development of transportation systems, including the addition of automatic road navigation, unmanned vehicles, and urban planning, which are important in both industry and daily living [1]. Automatic road extraction from high-resolution optical remote sensing imagery is a fundamental task [2]. Road extraction from high-resolution images has two challenges: (i) The images have complex road structures; remote sensing images are usually characterized by complexity in the form of heterogeneous regions with large intraclass variations and lower interclass variations [3]. (ii) The objects in images are blocked by obstacles, either through shadow occlusion or visual occlusion. Roads can be roughly recognized from images with shadow occlusion, but those with visual occlusion cannot reflect road information.

Methods

Results

Discussion

Conclusion