A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery
Hyperspectral remote sensing can obtain both spatial and spectral information of ground objects. It is an important prerequisite for a hyperspectral remote sensing application to make good use of spectral and image features. Therefore, we improved the Convolutional Neural Network (CNN) model by extr...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/13/15/2927 |
_version_ | 1797525162414833664 |
---|---|
author | Chengming Ye Hongfu Li Chunming Li Xin Liu Yao Li Jonathan Li Wesley Nunes Gonçalves José Marcato Junior |
author_facet | Chengming Ye Hongfu Li Chunming Li Xin Liu Yao Li Jonathan Li Wesley Nunes Gonçalves José Marcato Junior |
author_sort | Chengming Ye |
collection | DOAJ |
description | Hyperspectral remote sensing can obtain both spatial and spectral information of ground objects. It is an important prerequisite for a hyperspectral remote sensing application to make good use of spectral and image features. Therefore, we improved the Convolutional Neural Network (CNN) model by extracting interior-edge-adjacency features of building roof and proposed a new CNN model with a flexible structure: Building Roof Identification CNN (BRI-CNN). Our experimental results demonstrated that the BRI-CNN can not only extract interior-edge-adjacency features of building roof, but also change the weight of these different features during the training process, according to selected samples. Our approach was tested using the Indian Pines (IP) data set and our comparative study indicates that the BRI-CNN model achieves at least 0.2% higher overall accuracy than that of the capsule network model, and more than 2% than that of CNN models. |
first_indexed | 2024-03-10T09:09:51Z |
format | Article |
id | doaj.art-1dcb0c4fef92444c89c20768c6ea406d |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T09:09:51Z |
publishDate | 2021-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-1dcb0c4fef92444c89c20768c6ea406d2023-11-22T06:06:20ZengMDPI AGRemote Sensing2072-42922021-07-011315292710.3390/rs13152927A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral ImageryChengming Ye0Hongfu Li1Chunming Li2Xin Liu3Yao Li4Jonathan Li5Wesley Nunes Gonçalves6José Marcato Junior7Key Laboratory of Earth Exploration and Information Technology of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Technology of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Technology of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Technology of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Mountain Hazards and Earth Surface Process, Chinese Academy of Sciences, Chengdu 610059, ChinaDepartments of Geography and Environmental Management and Systems Design Engineering, University of Waterloo, 200 University Ave West, Waterloo, ON N2L 3G1, CanadaFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Campo Grande 79070-900, Mato Grosso do Sul, BrazilFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Campo Grande 79070-900, Mato Grosso do Sul, BrazilHyperspectral remote sensing can obtain both spatial and spectral information of ground objects. It is an important prerequisite for a hyperspectral remote sensing application to make good use of spectral and image features. Therefore, we improved the Convolutional Neural Network (CNN) model by extracting interior-edge-adjacency features of building roof and proposed a new CNN model with a flexible structure: Building Roof Identification CNN (BRI-CNN). Our experimental results demonstrated that the BRI-CNN can not only extract interior-edge-adjacency features of building roof, but also change the weight of these different features during the training process, according to selected samples. Our approach was tested using the Indian Pines (IP) data set and our comparative study indicates that the BRI-CNN model achieves at least 0.2% higher overall accuracy than that of the capsule network model, and more than 2% than that of CNN models.https://www.mdpi.com/2072-4292/13/15/2927hyperspectral imagespectral and spatial featureConvolutional Neural Network (CNN)interior-edge-adjacency featuresbuilding roof |
spellingShingle | Chengming Ye Hongfu Li Chunming Li Xin Liu Yao Li Jonathan Li Wesley Nunes Gonçalves José Marcato Junior A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery Remote Sensing hyperspectral image spectral and spatial feature Convolutional Neural Network (CNN) interior-edge-adjacency features building roof |
title | A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery |
title_full | A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery |
title_fullStr | A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery |
title_full_unstemmed | A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery |
title_short | A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery |
title_sort | building roof identification cnn based on interior edge adjacency features using hyperspectral imagery |
topic | hyperspectral image spectral and spatial feature Convolutional Neural Network (CNN) interior-edge-adjacency features building roof |
url | https://www.mdpi.com/2072-4292/13/15/2927 |
work_keys_str_mv | AT chengmingye abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT hongfuli abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT chunmingli abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT xinliu abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT yaoli abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT jonathanli abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT wesleynunesgoncalves abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT josemarcatojunior abuildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT chengmingye buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT hongfuli buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT chunmingli buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT xinliu buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT yaoli buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT jonathanli buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT wesleynunesgoncalves buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery AT josemarcatojunior buildingroofidentificationcnnbasedoninterioredgeadjacencyfeaturesusinghyperspectralimagery |