MetaBoost: A Novel Heterogeneous DCNNs Ensemble Network With Two-Stage Filtration for SAR Ship Classification

Zheng, Hao; Hu, Zhigang; Liu, Jianjun; Huang, Yuhang; Zheng, Meiguang

doi:10.1109/lgrs.2022.3180793

Cited by 15 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dong et al [13] designed a deeper SAR ship classification model by introducing a residual module. Zheng [14] proposed an ensemble network to improve the robustness and accuracy of classification by fusing multiple heterogeneous deep convolutional neural networks. Huang et al [15] presented a novel method for CNNs, called Group Squeeze Stimulated Sparsely Connected Convolutional Networks (GSESCNNs), which made the concatenation of feature maps from different layers more efficient through sparse connection operations.…”

Section: B Modern Deep Feature Methodsmentioning

confidence: 99%

“…To evaluate the feasibility and effectiveness of MFCFNet to fuse handcrafted features, we perform extensive experimental analysis on two popular SAR ship datasets like other scholars [29][30][31]. The distribution ratio and preprocessing of the datasets are the same as our previous work [14]. Table I lists the distribution of two datasets, such as categories, totals, and allocations.…”

Section: A Data Descriptionmentioning

confidence: 99%

See 1 more Smart Citation

Multifeature Collaborative Fusion Network With Deep Supervision for SAR Ship Classification

Zheng

Liu

et al. 2023

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

Multi-feature SAR ship classification aims to build models that can process, correlate, and fuse information from handcrafted features and deep features. Although handcrafted ones provide rich expert knowledge, current fusion methods do not thoroughly investigate the relatively important of handcrafted features with deep features, feature contribution imbalance, and the way features learn collaboratively. In this paper, a novel multi-feature collaborative fusion network with deep supervision (MFCFNet) is proposed to effectively realize handcrafted feature and deep feature fusion in SAR ship classification task. Specifically, our framework mainly includes two types of feature extraction branches, a knowledge supervision and collaboration module, and a feature fusion and contribution assignment module. The former module improves the feature map quality learned by each branch through auxiliary feature supervision, and introduces synergy loss to facilitate the information interaction between deep features and handcrafted features. The latter utilizes an attention mechanism to adaptively balance the importance among various features, and to assign the corresponding feature contribution to the total loss function based on the generated feature weights. We conduct extensive experimental and ablation studies on two public OpenSARShip-1.0 and FUSAR-Ship datasets, and the results show that MFCFNet is effective and outperforms single deep feature and multi-feature models based on previous Internal FC Layer and Terminal FC Layer fusion, and exhibits better performance than the current state-of-the-art methods.

show abstract

Section: B Modern Deep Feature Methodsmentioning

confidence: 99%

Section: A Data Descriptionmentioning

confidence: 99%

Multifeature Collaborative Fusion Network With Deep Supervision for SAR Ship Classification

Zheng

Liu

et al. 2023

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, due to its end-to-end learning method, deep learning can learn feature information highly relevant to the task and has made remarkable achievements in the problem of PolSAR image classification, such as stacked autoencoders, 13 recurrent neural networks, 14 deep belief networks, 15 and deep convolutional neural networks. 16 Among them, CNN can well obtain the spatial features of images by virtue of its unique convolutional network structure. Therefore, the method based on convolutional neural network (CNN) has achieved remarkable results in PolSAR image classification.…”

Section: Introductionmentioning

confidence: 99%

“…Deep learning technology has seen extensive utilization across diverse machine learning domains in recent years, owing to its continuous advancements. Moreover, due to its end-to-end learning method, deep learning can learn feature information highly relevant to the task and has made remarkable achievements in the problem of PolSAR image classification, such as stacked autoencoders, 13 recurrent neural networks, 14 deep belief networks, 15 and deep convolutional neural networks 16 …”

Section: Introductionmentioning

confidence: 99%

Multi-scale contrastive learning method for PolSAR image classification

Hua,

Wang,

Sun

et al. 2024

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Although deep learning-based methods have made remarkable achievements in polarimetric synthetic aperture radar (PolSAR) image classification, these methods require a large number of labeled samples. However, for PolSAR image classification, it is difficult to obtain a large number of labeled samples, which requires extensive human labor and material resources. Therefore, a new PolSAR image classification method based on multi-scale contrastive learning is proposed, which can achieve good classification results with only a small number of labeled samples. During the pre-training process, we propose a multi-scale contrastive learning network model that uses the characteristics of the data itself to train the network by contrastive training. In addition, to capture richer feature information, a multi-scale network structure is introduced. In the training process, considering the diversity and complexity of PolSAR images, we design a hybrid loss function combining the supervised and unsupervised information to achieve better classification performance with limited labeled samples. The experimental results on three real PolSAR datasets have demonstrated that the proposed method outperforms other comparison methods, even with limited labeled samples.

show abstract

“…the effectiveness of these methods. This issue underscores a critical disconnect between the theoretical design of ATR methods and their practical applications, with existing methods often falling short when deployed in real-world contexts [17]- [19]. This problem, termed as SAR ATR with limited training data, has recently become a focus in research [20]- [23].…”

Section: Introductionmentioning

confidence: 99%