Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks
The interpretation of land use and land cover (LULC) is an important issue in the fields of high-resolution remote sensing (RS) image processing and land resource management. Fully training a new or existing convolutional neural network (CNN) architecture for LULC classification requires a large amo...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-11-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/10/12/1889 |
id |
doaj-53f60680ede4473fb2305e2dee18ad95 |
---|---|
record_format |
Article |
spelling |
doaj-53f60680ede4473fb2305e2dee18ad952020-11-25T00:37:30ZengMDPI AGRemote Sensing2072-42922018-11-011012188910.3390/rs10121889rs10121889Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural NetworksBin Zhang0Cunpeng Wang1Yonglin Shen2Yueyan Liu3School of Geography and Information Engineering, China University of Geoscience, Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geoscience, Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geoscience, Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geoscience, Wuhan 430074, ChinaThe interpretation of land use and land cover (LULC) is an important issue in the fields of high-resolution remote sensing (RS) image processing and land resource management. Fully training a new or existing convolutional neural network (CNN) architecture for LULC classification requires a large amount of remote sensing images. Thus, fine-tuning a pre-trained CNN for LULC detection is required. To improve the classification accuracy for high resolution remote sensing images, it is necessary to use another feature descriptor and to adopt a classifier for post-processing. A fully connected conditional random fields (FC-CRF), to use the fine-tuned CNN layers, spectral features, and fully connected pairwise potentials, is proposed for image classification of high-resolution remote sensing images. First, an existing CNN model is adopted, and the parameters of CNN are fine-tuned by training datasets. Then, the probabilities of image pixels belong to each class type are calculated. Second, we consider the spectral features and digital surface model (DSM) and combined with a support vector machine (SVM) classifier, the probabilities belong to each LULC class type are determined. Combined with the probabilities achieved by the fine-tuned CNN, new feature descriptors are built. Finally, FC-CRF are introduced to produce the classification results, whereas the unary potentials are achieved by the new feature descriptors and SVM classifier, and the pairwise potentials are achieved by the three-band RS imagery and DSM. Experimental results show that the proposed classification scheme achieves good performance when the total accuracy is about 85%.https://www.mdpi.com/2072-4292/10/12/1889remote sensingimage classificationfully connected conditional random fields (FC-CRF)convolutional neural networks (CNN) |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Bin Zhang Cunpeng Wang Yonglin Shen Yueyan Liu |
spellingShingle |
Bin Zhang Cunpeng Wang Yonglin Shen Yueyan Liu Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks Remote Sensing remote sensing image classification fully connected conditional random fields (FC-CRF) convolutional neural networks (CNN) |
author_facet |
Bin Zhang Cunpeng Wang Yonglin Shen Yueyan Liu |
author_sort |
Bin Zhang |
title |
Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks |
title_short |
Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks |
title_full |
Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks |
title_fullStr |
Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks |
title_full_unstemmed |
Fully Connected Conditional Random Fields for High-Resolution Remote Sensing Land Use/Land Cover Classification with Convolutional Neural Networks |
title_sort |
fully connected conditional random fields for high-resolution remote sensing land use/land cover classification with convolutional neural networks |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2018-11-01 |
description |
The interpretation of land use and land cover (LULC) is an important issue in the fields of high-resolution remote sensing (RS) image processing and land resource management. Fully training a new or existing convolutional neural network (CNN) architecture for LULC classification requires a large amount of remote sensing images. Thus, fine-tuning a pre-trained CNN for LULC detection is required. To improve the classification accuracy for high resolution remote sensing images, it is necessary to use another feature descriptor and to adopt a classifier for post-processing. A fully connected conditional random fields (FC-CRF), to use the fine-tuned CNN layers, spectral features, and fully connected pairwise potentials, is proposed for image classification of high-resolution remote sensing images. First, an existing CNN model is adopted, and the parameters of CNN are fine-tuned by training datasets. Then, the probabilities of image pixels belong to each class type are calculated. Second, we consider the spectral features and digital surface model (DSM) and combined with a support vector machine (SVM) classifier, the probabilities belong to each LULC class type are determined. Combined with the probabilities achieved by the fine-tuned CNN, new feature descriptors are built. Finally, FC-CRF are introduced to produce the classification results, whereas the unary potentials are achieved by the new feature descriptors and SVM classifier, and the pairwise potentials are achieved by the three-band RS imagery and DSM. Experimental results show that the proposed classification scheme achieves good performance when the total accuracy is about 85%. |
topic |
remote sensing image classification fully connected conditional random fields (FC-CRF) convolutional neural networks (CNN) |
url |
https://www.mdpi.com/2072-4292/10/12/1889 |
work_keys_str_mv |
AT binzhang fullyconnectedconditionalrandomfieldsforhighresolutionremotesensinglanduselandcoverclassificationwithconvolutionalneuralnetworks AT cunpengwang fullyconnectedconditionalrandomfieldsforhighresolutionremotesensinglanduselandcoverclassificationwithconvolutionalneuralnetworks AT yonglinshen fullyconnectedconditionalrandomfieldsforhighresolutionremotesensinglanduselandcoverclassificationwithconvolutionalneuralnetworks AT yueyanliu fullyconnectedconditionalrandomfieldsforhighresolutionremotesensinglanduselandcoverclassificationwithconvolutionalneuralnetworks |
_version_ |
1725301054455676928 |