A Deep Neural Network Combined CNN and GCN for Remote Sensing Scene Classification

Liang, Jiali; Deng, Yufan; Zeng, Dan

doi:10.1109/jstars.2020.3011333

Cited by 100 publications

(65 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CM of 50% of training images is shown in Figure 3 to analyse the results in detail. All the categories are predicted correctly, and only the classes of 'aeroplane','building', 'harbour', and 'sparseresidential' have some difficult samples is compared with MG-CAP [36], RTN [42], DCNN [37], MSCP [38], Combined CNN and GCN [39], MSITL [40], ADFF [41], and MSDFF [35]. Among all SOTAs, the proposed ERANet has achieved the best score with 20% and 10% training set.…”

Section: Resultsmentioning

confidence: 99%

Efficient recurrent attention network for remote sensing scene classification

Liang¹,

Wang

2021

IET Image Processing

View full text Add to dashboard Cite

Scene classification for remote sensing is a popular topic, and many recent convolutional neural networks (CNNs)-based methods have shown the great model capacity and learning ability of highly discriminative features. Given a large number of training data, CNN can extract extensive features and learn to predict a remote sensing image. However, for supervised learning tasks, deep models often rely on a large number of labelled remote sensing images, which are difficult to pre-process. Thus, training a lightweight deep learning model is essential. Easy-classified and hard samples may also cause an imbalance of training set and lead the model to overwhelm the loss function. Accordingly, a novel Efficient Recurrent Attention Network (ERANet) for remote sensing scene classification is proposed. Different from traditional deep learning methods, Efficientnet-B0 is introduced as a lightweight backbone for the ARCNet framework, replacing the original one. By applying the modified efficient backbone, the low Floating Point Operations (FLOPs) and parameter numbers of the proposed ERANet are maintained. The significance of focal loss is determined and applied to address the sample imbalance problem and yield a desirable performance. Extensive experiments on several challenging remote sensing scene classification data sets prove the efficiency of the proposed ERANet.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

Section: Resultsmentioning

confidence: 99%

Efficient recurrent attention network for remote sensing scene classification

Liang¹,

Wang

2021

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Due to their convincing performance and high interpretability, GNNs have recently been widely applied in a wide range of problem domains across scene graph generation, scene classification (Gao, Shi, & Li et al, 2021;Liang, Deng, & Zeng, 2020), point clouds segmentation/classification (Wen, Li, Yao, Peng, & Chi, 2021;Widyaningrum, Bai, Fajari, & Lindenbergh, 2021), text classification, traffic forecasting (Peng, Wang, Du, Bhuiyan, Ma, Liu & Yu, 2020), and event detection (Guo et al, 2019). However, applications of GNNs in RS are just beginning.…”

Section: Graph Neural Networkmentioning

confidence: 99%

Deep learning for processing and analysis of remote sensing big data: a technical review

2021

View full text Add to dashboard Cite

In recent years, the rapid development of Earth observation technology has produced an increasing growth in remote sensing big data, posing serious challenges for effective and efficient processing and analysis. Meanwhile, there has been a massive rise in deeplearning-based algorithms for remote sensing tasks, providing a large opportunity for remote sensing big data. In this article, we initially summarize the features of remote sensing big data. Subsequently, following the pipeline of remote sensing tasks, a detailed and technical review is conducted to discuss how deep learning has been applied to the processing and analysis of remote sensing data, including geometric and radiometric processing, cloud masking, data fusion, object detection and extraction, landuse/cover classification, change detection and multitemporal analysis. Finally, we discussed technical challenges and concluded directions for future research in deep-learning-based applications for remote sensing big data.

show abstract

“…The authors utilized the complementarity between local and global features by exploiting the features of different scales and discarding the redundant information in features. Liang et al [37] introduced a novel two-stream architecture combining global-based visual features and object-based features. The model first extracts the appearance visual features from the scene image using convolutional neural network and later detects the ground objects and finally constructs a graph to learn spatial features using a graph convolutional network.…”

Section: Related Workmentioning

confidence: 99%

“…Therefore, using a CNN as a "black box" to classify remote sensing images may be not good enough for complex scenes. Even though several works [44,6] attempted to focus on the critical local image patches and discard the useless information, they still only utilize the visual information [37].…”

mentioning

confidence: 99%

A Discriminative Neighborhood-Based Collaborative Learning for Remote Sensing Scene Classification

Usman¹,

Hoque²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

The bag-of-words (BoW) model is one of the most popular representation methods for image classification. However, the lack of spatial information, change of illumination, and inter-class similarity among scene categories impair its performance in the remote-sensing domain. To alleviate these issues, this paper proposes to explore the spatial dependencies between different image regions and introduce a neighborhood-based collaborative learning (NBCL) for remote-sensing scene classification. Particularly, our proposed method employs multilevel features learning based on small, medium, and large neighborhood regions to enhance the discriminative power of image representation. To achieve this, image patches are selected through a fixed-size sliding window where each image is represented by four independent image region sequences. Apart from multilevel learning, we explicitly impose Gaussian pyramids to magnify the visual information of the scene images and optimize their position and scale parameters locally. Motivated by this, a local descriptor is exploited to extract multilevel and multiscale features that we represent in terms of codewords histogram by performing k-means clustering. Finally, a simple fusion strategy is proposed to balance the contribution of these features, and the fused features are incorporated into a Bidirectional Long Short-Term Memory (BiLSTM) network for constructing the final representation for classification. Experimental results on NWPU-RESISC45, AID, UC-Merced, and WHU-RS datasets demonstrate that the proposed approach not only surpasses the conventional bag-of-words approaches but also yields significantly higher classification performance than the existing state-of-the-art deep learning methods used nowadays.

show abstract

A Deep Neural Network Combined CNN and GCN for Remote Sensing Scene Classification

Cited by 100 publications

References 51 publications

Efficient recurrent attention network for remote sensing scene classification

Efficient recurrent attention network for remote sensing scene classification

Deep learning for processing and analysis of remote sensing big data: a technical review

A Discriminative Neighborhood-Based Collaborative Learning for Remote Sensing Scene Classification

Contact Info

Product

Resources

About