Few-Shot Learning Neural Network for SAR Target Recognition

Lu, Da; Cao, Lihua; Liu, Hongwei

doi:10.1109/apsar46974.2019.9048517

Cited by 11 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep transfer learning has also been applied to fewshot SAR image classification [49]. For the identification of few-shot SAR targets, Lu et al presented a deep convolutional neural network based on transfer learning [50]. They utilized the refined model to categorize new cases after fine-tuning a pre-trained network model on a few samples from new classes.…”

Section: B Few-shot Classification In Sar Imagesmentioning

confidence: 99%

“…By leveraging prior knowledge or metalearning, few-shot learning algorithms can effectively generalize to new tasks with limited labeled data [14]. In the context of SAR image classification, few-shot learning approaches have been proposed to extract and leverage the underlying features of SAR images, which can be used to classify images with a limited number of labeled examples [15]. Recently, various few-shot learning methods have been developed, including data augmentation, meta-learning, and transfer learning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Scattering-Point Topology for Few-Shot Ship Classification in SAR Images

Zhang,

Lu,

Qiu

et al. 2023

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

View full text Add to dashboard Cite

Synthetic aperture radar (SAR) has emerged as a critical technology for detecting and classifying objects such as ships in challenging environments. However, few-shot learning remains challenging due to the limited availability of labeled SAR data, complex radar backscatter, and variations in imaging parameters. In this paper, we propose a novel network, scattering point topology for few-shot ship classification (SPT-FSC), which addresses these challenges by incorporating scattering characteristics into the network learning process through a scattering point topology (SPT) based on scattering key points. We design a topology encoding branch (TEB) through a series of operations to encode the topological information of scattering points, resulting in a SPT embedding that improves the network's adaptability to the imaging mechanism and reduces imaging variability in SAR images. To effectively fuse the SPT embedding and image features extracted from a convolutional neural network (CNN), we introduce a novel mechanism named reciprocal feature fusion attention (RFFA). Additionally, to address the limited diversity in the training data, we apply fine-tuning based methodologies and construct a fine-grained ship classification dataset by combining the OpenSARShip and FUSAR-Ship datasets. Our comprehensive experiments on these datasets demonstrate the effectiveness of our proposed SPT-FSC method, achieving high accuracy and robustness in few-shot ship classification tasks for SAR images, outperforming existing methods in this domain.

show abstract

Section: B Few-shot Classification In Sar Imagesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Scattering-Point Topology for Few-Shot Ship Classification in SAR Images

Zhang,

Lu,

Qiu

et al. 2023

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

View full text Add to dashboard Cite

show abstract

“…However, for data collection of noncooperative targets, it is common to face the challenge of limited labeled SAR data. Some scholars have focused on solving the problem of few-shot SAR target recognition [16]- [19], [40]- [42]. Siamese neural network, whose target category was outputted by the classifier [18] and not by the similarity discriminator [43], was improved to solve the challenges of few-shot SAR target recognition.…”

Section: Related Workmentioning

confidence: 99%

“…Siamese neural network, whose target category was outputted by the classifier [18] and not by the similarity discriminator [43], was improved to solve the challenges of few-shot SAR target recognition. Instead of cross-entropy loss, the triple loss was introduced to the deep learning framework for SAR image recognition with the experiment configuration about only one few-shot class and nine classes with sufficient data [16]. A meta-learning framework named MSAR [41], consisting of a meta-learner and a base learner, can learn a good initialization as well as a proper update strategy.…”

Section: Related Workmentioning

confidence: 99%

“…The use of few-shot learning purpose is growing in the fields of computer vision and pattern recognition. However, only a few methods [16]- [20], including transfer learning [21] and metric or loss based improvement, are used in SAR image interpretation. Unfortunately, these methods emphasize the improvements of models and networks but overlook the mechanism, which contains SAR domain knowledge, in SAR targets.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Domain Knowledge Powered Two-Stream Deep Network for Few-Shot SAR Vehicle Recognition

Zhang

Leng

Feng

et al. 2022

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Synthetic aperture radar (SAR) target recognition faces the challenge that there are very little labeled data. Although few-shot learning methods are developed to extract more information from a small amount of labeled data to avoid overfitting problems, recent few-shot or limited-data SAR target recognition algorithms overlook the unique SAR imaging mechanism. Domain knowledge-powered two-stream deep network (DKTS-N) is proposed in this study, which incorporates SAR domain knowledge related to the azimuth angle, the amplitude, and the phase data of vehicles, making it a pioneering work in few-shot SAR vehicle recognition. The two-stream deep network, extracting the features of the entire image and image patches, is proposed for more effective use of the SAR domain knowledge. To measure the structural information distance between the global and local features of vehicles, the deep Earth mover's distance is improved to cope with the features from a two-stream deep network. Considering the sensitivity of the azimuth angle in SAR vehicle recognition, the nearest neighbor classifier replaces the structured fully connected layer for K -shot classification. All experiments are conducted under the configuration that the SARSIM and the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset work as a source and target task, respectively. Our proposed DKTS-N achieved 49.26% and 96.15% under ten-way one-shot and ten-way 25-shot, whose labeled samples are randomly selected from the training set. In standard operating condition (SOC) as well as three extended operating conditions (EOCs), DKTS-N demonstrated overwhelming advantages in accuracy and time consumption compared with other few-shot learning methods in K -shot recognition tasks.

show abstract