Fusing Two Convolutional Neural Networks For High-Resolution Scene Classification
Keywords
convolutional neural network; Deep feature extraction; Feature learning; High-resolution image
Abstract
This paper presents a novel deep convolutional feature fusion (ConvFF) approach for high-resolution scene classification, characterizing the well-known deep convolutional neural network (ConvNet) approach. The proposed ConvFF approach starts by generating an initial feature representation of the original scenes under exploration from two deep ConvNets pre-trained on two different large amount of labeled data. After the pre-training phase, we fine tune the two deep ConvNets consisting of mainly objects and scenes respectively in a supervised manner using the target training images. Then we propose to fuse the extracted two types of convolutional features provided by the last fully-connected (FC) layer, respectively. Finally, the fused convolutional features are fed as input to a SVM classifier for classification. The proposed method is evaluated by using two challenging high-resolution scene datasets. Experimental results show that the proposed method can effectively extract complementary features of the scenes and capture local spatial patterns, consistently outperforming several state-of-the-art methods.
Publication Date
12-1-2017
Publication Title
International Geoscience and Remote Sensing Symposium (IGARSS)
Volume
2017-July
Number of Pages
3242-3245
Document Type
Article; Proceedings Paper
Personal Identifier
scopus
DOI Link
https://doi.org/10.1109/IGARSS.2017.8127688
Copyright Status
Unknown
Socpus ID
85041833740 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85041833740
STARS Citation
Bian, Xiaoyong; Chen, Chen; Sheng, Yuxia; Xu, Yan; and Du, Qian, "Fusing Two Convolutional Neural Networks For High-Resolution Scene Classification" (2017). Scopus Export 2015-2019. 7148.
https://stars.library.ucf.edu/scopus2015/7148