A Multi-Domain Collaborative Transfer Learning Method with Multi-Scale Repeated Attention Mechanism for Underwater Side-Scan Sonar Image Classification

Cheng, Zhen; Huo, Guanying; Li, Haisen

doi:10.3390/rs14020355

Cited by 26 publications

(8 citation statements)

References 43 publications

(84 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, in order to obtain a more accurate sonar image dataset of seabed targets, Cheng et al proposed a multi-domain cooperative transfer learning method with a multi-scale repeated attention mechanism based on the SIFT algorithm. It improves the accuracy of underwater sonar image classification [91].…”

Section: Applicationsmentioning

confidence: 99%

Sonar image intelligent processing in seabed pipeline detection: review and application

Shi,

Cao,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Subsea pipelines rely primarily on imaging sonar for detection and identification. We analyze the imaging principles of side scan sonar, multi-beam sonar, synthetic aperture sonar, seafloor penetrating sonar and forward-looking sonar. We discuss their effectiveness in detecting seabed pipelines, as well as their limitations in image recognition capabilities. As intelligent algorithms have become increasingly important in the field of image processing, we review the sonar image intelligent detection and recognition algorithms in the past six years and summarize the internal principles and application effects of classic algorithms such as SIFT (Scale-Invariant Feature Transform), KMA (K-means algorithm), and CFAR (Constant False-Alarm Rate) that currently show good application prospects. Simultaneously, we review the particular strengths exhibited by these algorithms, such as contour feature extraction, image segmentation and clustering, target recognition under background noise, etc. The research on intelligent processing of sonar images opens up a new way to solve the difficult problem of the seabed targets detection and recognition.

show abstract

Section: Applicationsmentioning

confidence: 99%

Sonar image intelligent processing in seabed pipeline detection: review and application

Shi,

Cao,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Moreover, disparities between composite and SSS image samples directly influence the target recognition outcome. Cheng et al [40] enhanced the noise resistance and recognition precision of the VGG-19 network by training its middle layer with synthetic aperture radar (SAR) data. Wang et al [41] employed ResNet-152 as the foundational network for target transfer recognition and introduced the location attention mechanism considering channel factors and the channel attention mechanism considering location factors.…”

Section: Related Workmentioning

confidence: 99%

An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism

Wang,

Li,

Dong

et al. 2023

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Recognizing targets through side-scan sonar (SSS) data by deep learning-based techniques has been particularly challenging. The primary challenge stems from the difficulty and time consumption associated with underwater acoustic data acquisition, which demands systematic explorations to obtain sufficient training samples for accurate deep learning-based models. Moreover, if the sample size of the available data is small, the design of effective target recognition models becomes complex. These challenges have posed significant obstacles to developing accurate SSS-based target recognition methods via deep learning models. However, utilizing multi-modal datasets to enhance the recognition performance of sonar images through knowledge transfer in deep networks appears promising. Owing to the unique statistical properties of various modal images, transitioning between different modalities can significantly increase the complexity of network training. This issue remains unresolved, directly impacting the target transfer recognition performance. To enhance the precision of categorizing underwater sonar images when faced with a limited number of mode types and data samples, this study introduces a crossed point-to-point second-order self-attention (PPCSSA) method based on double-mode sample transfer recognition. In the PPCSSA method, first-order importance features are derived by extracting key horizontal and longitudinal point-to-point features. Based on these features, the self-supervised attention strategy effectively removes redundant features, securing the second-order significant features of SSS images. This strategy introduces a potent low-mode-type small-sample learning method for transfer learning. Classification experiment results indicate that the proposed method excels in extracting key features with minimal training complexity. Moreover, experimental outcomes underscore that the proposed technique enhances recognition stability and accuracy, achieving a remarkable overall accuracy rate of 99.28%. Finally, the proposed method maintains high recognition accuracy even in noisy environments.

show abstract

“…Li et al [48] had proposed a transfer learning method for sonar image classification and target detection called the texture feature removal network, to deal with the problem of few targets in sonar images. Cheng et al [49] had proposed a multi-domain collaborative transfer learning (MDCTL) method with multi-scale repeated attention mechanism (MSRAM) for improving the accuracy of underwater sonar image classification.…”

Section: Target Detection For Sss Imagesmentioning

confidence: 99%

D2MFNet: a Dual-Domain Multi-Frequency Network for Side-Scan Sonar Image Detection

Wang,

Zhang,

et al. 2023

Preprint

View full text Add to dashboard Cite

The utilization of side-scan sonar (SSS) technology plays a pivotal role in ensuring the safety of underwater environments and facilitating the discovery of aquatic resources during daily production activities. Within this field, the use of SSS images for seabed target detection has gained widespread recognition. However, many existing detection approaches primarily concentrate on tracking the evolution path of optical image object detection tasks, resulting in complex structures and limited versatility. To tackle this issue, we introduce a pioneering Dual-Domain Multi-Frequency Network (D2MFNet) meticulously crafted to harness the distinct characteristics of SSS image detection. To maximize the information extracted from SSS images across various frequency ranges, we introduce a novel method for calculating the information flow of target ranges and propose a Multi-Frequency Combined Attention Mechanism (MFCAM). This mechanism amplifies the relevance of dual-domain features across different channels. Moreover, recognizing that SSS images can provide richer insights after frequency domain conversion, we introduce a Dual-Domain Feature Pyramid Network D2FPN. By incorporating frequency domain information representation, D2FPN significantly augments the depth and breadth of feature information in underwater small datasets. Our methods are seamlessly designed for integration into existing networks, offering plug-and-play functionality with substantial performance enhancements. We have conducted extensive experiments to validate the efficacy of our proposed techniques, and the results showcase their state-of-the-art performance. We will make our code and models publicly available at https://dagshub.com/estrellaww00/D2MFNet.

show abstract

A Multi-Domain Collaborative Transfer Learning Method with Multi-Scale Repeated Attention Mechanism for Underwater Side-Scan Sonar Image Classification

Cited by 26 publications

References 43 publications

Sonar image intelligent processing in seabed pipeline detection: review and application

Sonar image intelligent processing in seabed pipeline detection: review and application

An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism

D2MFNet: a Dual-Domain Multi-Frequency Network for Side-Scan Sonar Image Detection

Contact Info

Product

Resources

About