Improving Deep Subdomain Adaptation by Dual-Branch Network Embedding Attention Module for SAR Ship Classification

Zhao, Shuangmei; Lang, Haitao

doi:10.1109/jstars.2022.3206753

Cited by 7 publications

(5 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DSAN: A migration learning fault diagnosis method that uses a deep new neural network to extract the target area data and the source area data and align in that feature space the features of both areas [73].…”

Section: Other Model Comparsion Experimentsmentioning

confidence: 99%

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

Wang,

Zhang

2024

IEEE Access

View full text Add to dashboard Cite

To address the problem that uneven sample distribution can affect the accuracy and stability of fault diagnosis outcomes, we propose a deep transfer learning-Res2Net-convolutional block attention mechanism model. Firstly, the deep migration technique is used to transfer weights of the imbalanced source domain samples to the balanced target domain, expanding the data samples. Secondly, in the feature extraction and detection phase, All eight residual blocks are embedded in convolution block attention to ensure that interference signals are suppressed and that the key fault features are retained. Multilayer feature fusion extracts faulty sample features from multiple residual blocks of the network, which are combined into the feature fusion layer by parallel splicing. Finally, this model is experimentally validated using two different bearing datasets, and the combined evaluation indexes of the new model under the severe imbalance condition are 95.80% and 95.85%, respectively, demonstrating the feasibility and excellent performance of the model.

show abstract

Section: Other Model Comparsion Experimentsmentioning

confidence: 99%

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

Wang,

Zhang

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The fine-grained ship classification task aims to classify ships into specific categories, such as bulk carriers, container ships, oil tankers, or even different types of warships [10], [15], [16], [21], [32], [33]. To address the unique challenges posed by fine-grained ship classification in remote sensing images, researchers have conducted numerous studies across three main areas:…”

Section: A Fine-grained Ship Classificationmentioning

confidence: 99%

“…Zhao et al created a dual-branch network that utilizes ResNet-50 as the deep branch and ResNet-18 as the shallow branch to enhance the performance of the subdomain adaptation. To extract more discriminative deep transferable features, they also embedded the Convolutional Block Attention Module (CBAM) after the first and last convolutional layer of each branch [21]. He et al proposed a Group of Bilinear Convolutional Neural Network (GBCNN) to extract discriminative ship representations from the pairwise vertical-horizontal polarization and verticalvertical polarization SAR images [27].…”

Section: A Fine-grained Ship Classificationmentioning

confidence: 99%

See 1 more Smart Citation

A Unified Multiple Proxy Deep Metric Learning Framework Embedded With Distribution Optimization for Fine-Grained Ship Classification in Remote Sensing Images

Xu,

Lang

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Improving ship classification performance in remote sensing imagery by deep metric learning (DML) is a newly emerging research topic and has good application prospects. From the perspective of the use of metric loss (classification loss and pairwise loss) and the way of proxy learning (a single proxy or multiple proxies), this study summarizes the existing DML methods into four representative frameworks and proposes a novel framework, namely, a Unified Multiple Proxy deep metric learning framework embedded with Distribution optimization (UMP+D). Specifically, the UMP+D not only unifies the combination of classification loss and pairwise loss into a single loss function containing only pairwise representation but also fuses it with multiple proxy learning. In addition, a distribution loss (DL) branch is embedded in the UMP+D to refine the distribution of samples in the feature embedding space to further tighten the intraclass samples and pull apart the interclass samples. Extensive experiments on two optical remote sensing datasets and one synthetic aperture radar (SAR) dataset demonstrate that the proposed UMP+D framework outperforms the existing frameworks and achieves state-of-theart performance.

show abstract

“…For feature extraction and classification, Chen et al [9] used a stacked autoencoder (SAE). Other networks, such as the deep belief network (DBN) [10], have been suggested for HSIs classification. Restricted Boltzmann machines are stacked to create DBNs, which are based on multivariate optical sensors.…”

Section: Introductionmentioning

confidence: 99%

Improved Three-Dimensional Inception Networks for Hyperspectral Remote Sensing Image Classification

Zhang

2023

IEEE Access

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) have been applied to the categorization of hyperspectral images. Typically, CNNs employ fixed convolution kernel sizes. This may be complicated by the high-dimensional features and the spatial-spectral features variability of hyperspectral data. In such situation, dealing with various dataset types flexibly is hampered by the fixed kernel size. Therefore, we chose a parallel design, an upgraded three-dimensional (3D) inception network, for hyperspectral image classification. Additionally, the adaptive band selection of hyperspectral remote sensing data employs dimensionality reduction based on interactive information entropy, allowing the selection of low-redundant bands with more abundant and discriminative information. Experimental results based on the publicly available hyperspectral datasets demonstrated that the enhanced 3D inception network can perform well with limited training samples. Moreover, the combinations of different dimensional convolutions in the 3D-2D and 3D-2D-1D inception networks achieved comparable classification results. According to the experimental findings, the suggested networks are generalised classification models with high classification rates.

show abstract

Improving Deep Subdomain Adaptation by Dual-Branch Network Embedding Attention Module for SAR Ship Classification

Cited by 7 publications

References 51 publications

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

A Unified Multiple Proxy Deep Metric Learning Framework Embedded With Distribution Optimization for Fine-Grained Ship Classification in Remote Sensing Images

Improved Three-Dimensional Inception Networks for Hyperspectral Remote Sensing Image Classification

Contact Info

Product

Resources

About