2018
DOI: 10.1155/2018/8639367
|View full text |Cite
|
Sign up to set email alerts
|

A Two-Stream Deep Fusion Framework for High-Resolution Aerial Scene Classification

Abstract: One of the challenging problems in understanding high-resolution remote sensing images is aerial scene classification. A well-designed feature representation method and classifier can improve classification accuracy. In this paper, we construct a new two-stream deep architecture for aerial scene classification. First, we use two pretrained convolutional neural networks (CNNs) as feature extractor to learn deep features from the original aerial image and the processed aerial image through saliency detection, re… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
75
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
7

Relationship

0
7

Authors

Journals

citations
Cited by 114 publications
(75 citation statements)
references
References 47 publications
0
75
0
Order By: Relevance
“…As can be seen from Table 3, our classification method, by fusing the GCFs and LOFs, achieved the highest overall accuracy of 96.85% and 92.81% using 50% and 20% training ratios, respectively. Worthy of mention is that our architecture outperformed the second-best model [51], which uses a feature fusion method to reconstruct global feature representation, with increases in the overall accuracy of 2.27% and 0.16%. The good performance of our method is mainly the results of the fusion of GCFs and LOFs.…”
Section: Classification Of Aidmentioning
confidence: 95%
See 1 more Smart Citation
“…As can be seen from Table 3, our classification method, by fusing the GCFs and LOFs, achieved the highest overall accuracy of 96.85% and 92.81% using 50% and 20% training ratios, respectively. Worthy of mention is that our architecture outperformed the second-best model [51], which uses a feature fusion method to reconstruct global feature representation, with increases in the overall accuracy of 2.27% and 0.16%. The good performance of our method is mainly the results of the fusion of GCFs and LOFs.…”
Section: Classification Of Aidmentioning
confidence: 95%
“…We utilized the open-source Caffe framework [49] to implement our proposed architecture. In the experiments, two training ratios are adopted for each dataset, following the work of [5,50,51] for a fair comparison. For the AID and RSSCN7 datasets, 50% and 20% of the samples are randomly selected as the training samples and the left for testing.…”
Section: Implementation Detailsmentioning
confidence: 99%
“…CNN is one of the most popular deep learning model in image classification and image recognition tasks . They are also used in solving many other image or text based machine learning problems –. Usually grid like representation of data such as images in pixels, words in a text document are used as input features.…”
Section: Deep Learning Models For Sentiment Classificationmentioning
confidence: 99%
“…[49][50][51] They are also used in solving many other image or text based machine learning problems. [52][53][54] Usually grid like representation of data such as images in pixels, words in a text document are used as input features.…”
Section: Convolutional Neural Network (Cnn)mentioning
confidence: 99%
“…The technique only provides high classification accuracy when small training sets were used. A two-stream deep fusion approach was introduced in [3] for categorizing the aerial scene with high-quality remote sensing images using ELM classifier with fused features. Though the approach provides better classification results, the false positive rate was not minimized.…”
Section: Introductionmentioning
confidence: 99%