JointNet: A Common Neural Network for Road and Building Extraction

Zhengxin Zhang,Yunhong Wang

doi:10.3390/rs11060696

Abstract

Automatic extraction of ground objects is fundamental for many applications of remote sensing. It is valuable to extract different kinds of ground objects effectively by using a general method. We propose such a method, JointNet, which is a novel neural network to meet extraction requirements for both roads and buildings. The proposed method makes three contributions to road and building extraction: (1) in addition to the accurate extraction of small objects, it can extract large objects with a wide receptive field. By switching the loss function, the network can effectively extract multi-type ground objects, from road centerlines to large-scale buildings. (2) This network module combines the dense connectivity with the atrous convolution layers, maintaining the efficiency of the dense connection connectivity pattern and reaching a large receptive field. (3) The proposed method utilizes the focal loss function to improve road extraction. The proposed method is designed to be effective on both road and building extraction tasks. Experimental results on three datasets verified the effectiveness of JointNet in information extraction of road and building objects.

Highlights

Automatic extraction of ground objects based on remote sensing images is an essential step in many applications, including urban planning, map services, automated driving services, business planning, change detection, etc
The ground truth 4.1o.2f. tNheastieodnaaltaLsaebtsorinatcolurydeosf ProaattderanreRaecseoggmnietinotnat(iNonLPaRnd) RcoeandteDrliantae.seInts our experiment, we evaluated theNLmPeRthroodads doantatsheetsswegemreebnutailttiobny Cdhaetansgete.t Wal.e[4u4s]e, dconthseist1ing~o1f82024 imimaaggees.s Tohfetghreoudnadtatsreutthasofthe thetsreaidnaintagsesetst,itnhcelu1d8e1s ro~a1d94areiamsaeggemseansttahtieovnaalinddatcioenntseertl,inaen.dInthoeu1r9e5xpe~r2im24ent,iwmeageevsaalusattheedtethsteing mestehto.ds on the segmentation dataset
There was no evidence that the network model trained with the focal loss function was superior to the model trained with the binary cross-entropy loss function on the road segmentation extraction task

Summary

Introduction

Automatic extraction of ground objects based on remote sensing images is an essential step in many applications, including urban planning, map services, automated driving services, business planning, change detection, etc. In these applications, the two most valuable parts are road and building information. There are many differences in image features between road and building objects. Buildings show differences in their color, shape and texture features due to differences in their function, design, and materials. It is difficult to design a general-purpose algorithm to extract all types of ground objects effectively based only on texture features and colors of images

Methods

Results

Conclusion