Object-Based Features for House Detection from RGB High-Resolution Images
Automatic building extraction from satellite images, an open research topic in remote sensing, continues to represent a challenge and has received substantial attention for decades. This paper presents an object-based and machine learning-based approach for automatic house detection from RGB high-re...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-03-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | http://www.mdpi.com/2072-4292/10/3/451 |
id |
doaj-9800741f8c77488d883623bed79a0150 |
---|---|
record_format |
Article |
spelling |
doaj-9800741f8c77488d883623bed79a01502020-11-25T01:00:19ZengMDPI AGRemote Sensing2072-42922018-03-0110345110.3390/rs10030451rs10030451Object-Based Features for House Detection from RGB High-Resolution ImagesRenxi Chen0Xinhui Li1Jonathan Li2School of Earth Science and Engineering, Hohai University, Nanjing 211100, ChinaSchool of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, ChinaDepartment of Geography and Environmental Management, University of Waterloo, Waterloo, ON N2L3G1, CanadaAutomatic building extraction from satellite images, an open research topic in remote sensing, continues to represent a challenge and has received substantial attention for decades. This paper presents an object-based and machine learning-based approach for automatic house detection from RGB high-resolution images. The images are first segmented by an algorithm combing a thresholding watershed transformation and hierarchical merging, and then shadows and vegetation are eliminated from the initial segmented regions to generate building candidates. Subsequently, the candidate regions are subjected to feature extraction to generate training data. In order to capture the characteristics of house regions well, we propose two kinds of new features, namely edge regularity indices (ERI) and shadow line indices (SLI). Finally, three classifiers, namely AdaBoost, random forests, and Support Vector Machine (SVM), are employed to identify houses from test images and quality assessments are conducted. The experiments show that our method is effective and applicable for house identification. The proposed ERI and SLI features can improve the precision and recall by 5.6% and 11.2%, respectively.http://www.mdpi.com/2072-4292/10/3/451building extractionobject recognitionmachine learningimage segmentationfeature extractionclassification |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Renxi Chen Xinhui Li Jonathan Li |
spellingShingle |
Renxi Chen Xinhui Li Jonathan Li Object-Based Features for House Detection from RGB High-Resolution Images Remote Sensing building extraction object recognition machine learning image segmentation feature extraction classification |
author_facet |
Renxi Chen Xinhui Li Jonathan Li |
author_sort |
Renxi Chen |
title |
Object-Based Features for House Detection from RGB High-Resolution Images |
title_short |
Object-Based Features for House Detection from RGB High-Resolution Images |
title_full |
Object-Based Features for House Detection from RGB High-Resolution Images |
title_fullStr |
Object-Based Features for House Detection from RGB High-Resolution Images |
title_full_unstemmed |
Object-Based Features for House Detection from RGB High-Resolution Images |
title_sort |
object-based features for house detection from rgb high-resolution images |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2018-03-01 |
description |
Automatic building extraction from satellite images, an open research topic in remote sensing, continues to represent a challenge and has received substantial attention for decades. This paper presents an object-based and machine learning-based approach for automatic house detection from RGB high-resolution images. The images are first segmented by an algorithm combing a thresholding watershed transformation and hierarchical merging, and then shadows and vegetation are eliminated from the initial segmented regions to generate building candidates. Subsequently, the candidate regions are subjected to feature extraction to generate training data. In order to capture the characteristics of house regions well, we propose two kinds of new features, namely edge regularity indices (ERI) and shadow line indices (SLI). Finally, three classifiers, namely AdaBoost, random forests, and Support Vector Machine (SVM), are employed to identify houses from test images and quality assessments are conducted. The experiments show that our method is effective and applicable for house identification. The proposed ERI and SLI features can improve the precision and recall by 5.6% and 11.2%, respectively. |
topic |
building extraction object recognition machine learning image segmentation feature extraction classification |
url |
http://www.mdpi.com/2072-4292/10/3/451 |
work_keys_str_mv |
AT renxichen objectbasedfeaturesforhousedetectionfromrgbhighresolutionimages AT xinhuili objectbasedfeaturesforhousedetectionfromrgbhighresolutionimages AT jonathanli objectbasedfeaturesforhousedetectionfromrgbhighresolutionimages |
_version_ |
1725214100800143360 |