A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds

Wang, Yuanyuan; Wang, Chao; Zhang, Hong; Dong, Yingbo; Wei, Sisi

doi:10.3390/rs11070765

Cited by 375 publications

(220 citation statements)

References 39 publications

Supporting

Mentioning

220

Contrasting

Order By: Relevance

“…China's Gaofen-3 satellite SAR and Sentinel-1 SAR datasets as the main data sources used in this study, with a total usage of 102 Gaofen-3 images and 108 Sentinel-1 SAR images [23] Figure 3. In addition, labelling tools are used to label the vessel position and for classification.…”

Section: Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Detection of Small Ship Objects Using Anchor Boxes Cluster and Feature Pyramid Network Model for SAR Imagery

Chen

Liu

Zhong

et al. 2020

JMSE

View full text Add to dashboard Cite

The synthetic aperture radar (SAR) has a special ability to detect objects in any climate and weather conditions. Consequently, SAR images are widely used in maritime transportation safety and fishery law enforcement for maritime object detection. Currently, deep-learning models are being extensively used for the detection of objects from images. Among them, the feature pyramid network (FPN) uses pyramids for representing semantic information regardless of the scale and has an improved accuracy of object detection. It is also suitable for the detection of multiple small ship objects in SAR images. This study aims to resolve the problems associated with small-object and multi-object ship detection in complex scenarios e.g., when a ship nears the port, by proposing a detection method based on an optimized FPN model. The feature pyramid model is first embedded in a traditional region proposal network (RPN) and mapped into a new feature space for object identification. Subsequently, the k-means clustering algorithm based on the shape similar distance (SSD) measure is used to optimize the FPN. Initial anchor boxes and tests are created using the SAR ship dataset. Experimental results show that the proposed algorithm for object detection shows an accuracy of 98.62%. Compared with Yolo, the RPN based on VGG/ResNet, FPN based on VGG/ResNet, and other models in complex scenarios, the proposed model shows a higher accuracy rate and better overall performance.

show abstract

Section: Datasetmentioning

confidence: 99%

“…Gaofen-3 satellite SAR and Sentinel-1 SAR datasets as the main data sources used in this study, with a total usage of 102 Gaofen-3 images and 108 Sentinel-1 SAR images[23]. The SAR dataset includes 43,819 ship data slices.…”

mentioning

confidence: 99%

Detection of Small Ship Objects Using Anchor Boxes Cluster and Feature Pyramid Network Model for SAR Imagery

Chen

Liu

Zhong

et al. 2020

JMSE

View full text Add to dashboard Cite

show abstract

“…The dataset used in this paper is a SAR dataset for ship detection published by the Digital Earth Laboratory of the Aerospace Information Research Institute, Chinese Academy of Sciences [50]. There are 102 GaoFen-3 images and 108 Sentinel-1 images that are used to constructed the dataset.…”

Section: Dataset Descriptionmentioning

confidence: 99%

Enhanced Feature Extraction for Ship Detection from Multi-Resolution and Multi-Scene Synthetic Aperture Radar (SAR) Images

et al. 2019

View full text Add to dashboard Cite

Independent of daylight and weather conditions, synthetic aperture radar (SAR) images have been widely used for ship monitoring. The traditional methods for SAR ship detection are highly dependent on the statistical models of sea clutter or some predefined thresholds, and generally require a multi-step operation, which results in time-consuming and less robust ship detection. Recently, deep learning algorithms have found wide applications in ship detection from SAR images. However, due to the multi-resolution imaging mode and complex background, it is hard for the network to extract representative SAR target features, which limits the ship detection performance. In order to enhance the feature extraction ability of the network, three improvement techniques have been developed. Firstly, multi-level sparse optimization of SAR image is carried out to handle clutters and sidelobes so as to enhance the discrimination of the features of SAR images. Secondly, we hereby propose a novel split convolution block (SCB) to enhance the feature representation of small targets, which divides the SAR images into smaller sub-images as the input of the network. Finally, a spatial attention block (SAB) is embedded in the feature pyramid network (FPN) to reduce the loss of spatial information, during the dimensionality reduction process. In this paper, experiments on the multi-resolution SAR images of GaoFen-3 and Sentinel-1 under complex backgrounds are carried out and the results verify the effectiveness of SCB and SAB. The comparison results also show that the proposed method is superior to several state-of-the-art object detection algorithms.

show abstract

“…where ( )is the original source signal, �( )is the separated signal, is the sample index and N is the number of samples of the signal.A better performance of the matched filter is shown by a higher the value of PSNR with minimum value of RMSE.To demonstrate the performance of the proposed model, multispectral image sample data is considered. (d,f,g,i) are the cropped ship chips from the Gaofen-3 images with Ultrafine Strip (UFS) imaging mode (Wang, et al, 2019) In a multispectral dataset, the band information is reported as the centre wavelength value that represents the centre point value of the wavelengths. Simulations are performed on noisy mixed sample signal data on Matlab® R 7.9 on a core i7 2.2 GHz PC using the USFFT software package.…”

Section: Applications: Radar/satellite Signal Denoisingmentioning

confidence: 99%

Modelling curvelet based signal processing problems via wavelet analysis

2019

El Sawah, S. (Ed.) MODSIM2019, 23rd International Congress on Modelling and Simulation.

View full text Add to dashboard Cite

Until recently, the wavelet transform has been used for Mathematical analysis and signal processing problems. But it suffers from the disadvantage of poor directionality, which has undermined its usage in many applications. The curvelet transform is a new extension of the wavelet transform, which aims to deal with intersecting phenomena occurring along curved edges in 2-D signals/images. In this work, a strong relationship between the curvelet and wavelet transforms has been established. This version of a wavelet based curvelet has been exploited to develop a full-fledged analytical framework, presenting it as an extension of well established wavelets. Due to computational complexity, instead of applying a curvelet directly, the curvelet in terms of wavelets has been employed more conveniently in the proposed signal denoising model. Finally, the performance factor analysis is performed on multispectral sample radar image data to demonstrate the efficiency of the proposed model. Besides computational gain, the proposed model shows better performance than the other signal processing models. Proposed model is equally applicable to both pulse signals and digital images.

show abstract

A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds

Cited by 375 publications

References 39 publications

Detection of Small Ship Objects Using Anchor Boxes Cluster and Feature Pyramid Network Model for SAR Imagery

Detection of Small Ship Objects Using Anchor Boxes Cluster and Feature Pyramid Network Model for SAR Imagery

Enhanced Feature Extraction for Ship Detection from Multi-Resolution and Multi-Scene Synthetic Aperture Radar (SAR) Images

Modelling curvelet based signal processing problems via wavelet analysis

Contact Info

Product

Resources

About