Ship Detection and Feature Visualization Analysis Based on Lightweight CNN in VH and VV Polarization Images

Geng, Xiaomeng; Shi, Lei; Li, Pingxiang; Zhao, Lingli; Sun, Weidong; Zhao, Jinqi

doi:10.3390/rs13061184

Cited by 21 publications

(24 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The lowcomplexity and lightweight M-LeNet was once proven to be effective for ship detection in the nearshore area [20]. Thus, the M-LeNet model in [20] is improved in the present study as the baseline backbone target module and loss prediction module in active learning, as shown in Figure 3. The two convolution blocks of M-LeNet are selected as loss prediction modules.…”

mentioning

confidence: 91%

“…A two-stage ship detection method was proposed; however, the interpretability of the model was not addressed [19]. Then, the visual feature was analyzed, and the accuracy of small ships was improved by two-stage ship detection [20]. Although the above-mentioned models have been widely used in weak and small target detection, they have difficulty achieving satisfying performance due to the fact that most of the small-sized ships are nonmetallic fishing boats, and that generating strong dihedral angle scattering is hard due to their simple structure, material, and target wobble [21].…”

Section: Introductionmentioning

confidence: 99%

“…In this study, an improved two-stage ship detection method by active learning scheme with a small number of labeled sample data is proposed. To begin with, an exponential inverse cumulative distribution function [20] is employed to estimate the segmentation threshold and obtain candidate detection results. Then, the candidate detection results are optimized by the rule of boundary box distance.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Small-Sized Ship Detection Nearshore Based on Lightweight Active Learning Model with a Small Number of Labeled Data for SAR Imagery

et al. 2021

Self Cite

View full text Add to dashboard Cite

Marine ship detection by synthetic aperture radar (SAR) is an important remote sensing technology. The rapid development of big data and artificial intelligence technology has facilitated the wide use of deep learning methods in SAR imagery for ship detection. Although deep learning can achieve a much better detection performance than traditional methods, it is difficult to achieve satisfying performance for small-sized ships nearshore due to the weak scattering caused by their material and simple structure. Another difficulty is that a huge amount of data needs to be manually labeled to obtain a reliable CNN model. Manual labeling each datum not only takes too much time but also requires a high degree of professional knowledge. In addition, the land and island with high backscattering often cause high false alarms for ship detection in the nearshore area. In this study, a novel method based on candidate target detection, boundary box optimization, and convolutional neural network (CNN) embedded with active learning strategy is proposed to improve the accuracy and efficiency of ship detection in nearshore areas. The candidate target detection results are obtained by global threshold segmentation. Then, the strategy of boundary box optimization is defined and applied to reduce the noise and false alarms caused by island and land targets as well as by sidelobe interference. Finally, a lightweight CNN embedded with active learning scheme is used to classify the ships using only a small labeled training set. Experimental results show that the performance of the proposed method for small-sized ship detection can achieve 97.78% accuracy and 0.96 F1-score with Sentinel-1 images in complex nearshore areas.

show abstract

mentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Small-Sized Ship Detection Nearshore Based on Lightweight Active Learning Model with a Small Number of Labeled Data for SAR Imagery

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this case, the efficiency will be enhanced dramatically as long as the network is well trained. The convolutional neural network (CNN) is one of the most successful models in various computer vision fields [2,[13][14][15]. The key to its superiority lies in the way it uses local connections and shared weights.…”

Section: Introductionmentioning

confidence: 99%

Self-Matching CAM: A Novel Accurate Visual Explanation of CNNs for SAR Image Interpretation

et al. 2021

View full text Add to dashboard Cite

Synthetic aperture radar (SAR) image interpretation has long been an important but challenging task in SAR imaging processing. Generally, SAR image interpretation comprises complex procedures including filtering, feature extraction, image segmentation, and target recognition, which greatly reduce the efficiency of data processing. In an era of deep learning, numerous automatic target recognition methods have been proposed based on convolutional neural networks (CNNs) due to their strong capabilities for data abstraction and mining. In contrast to general methods, CNNs own an end-to-end structure where complex data preprocessing is not needed, thus the efficiency can be improved dramatically once a CNN is well trained. However, the recognition mechanism of a CNN is unclear, which hinders its application in many scenarios. In this paper, Self-Matching class activation mapping (CAM) is proposed to visualize what a CNN learns from SAR images to make a decision. Self-Matching CAM assigns a pixel-wise weight matrix to feature maps of different channels by matching them with the input SAR image. By using Self-Matching CAM, the detailed information of the target can be well preserved in an accurate visual explanation heatmap of a CNN for SAR image interpretation. Numerous experiments on a benchmark dataset (MSTAR) verify the validity of Self-Matching CAM.

show abstract

“…To detect a target against a complex background, the self-adaptive method based on the local variance weighted information entropy (VWIE) has been developed [9,10]. In recent years, a ship detection method based on convolutional neural networks (CNN) has been proposed [11,12]. The method is conventional and performs well.…”

Section: Introductionmentioning

confidence: 99%

Simulation of a Wideband Radar Echo of a Target on a Dynamic Sea Surface

Jiang

Wang

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Unlike a generally rough ground surface, the sea surface varies over time. To analyze the impact of the motion of sea waves on the synthetic aperture radar (SAR) image of a target, the wideband echo simulation method based on a frequency domain is used. For the wideband echo, the electromagnetic (EM) scattering properties of the main frequency components are analyzed by the simulation method. Based on the EM scattering properties, the echo can be accurately simulated by using the inverse fast Fourier transformation (IFFT). Combined with the flight path of the radar, the echo of each pulse can be simulated to obtain the SAR image. The correct evaluation of the EM scattering properties is indispensable to the acquisition of an accurate SAR image. For complex targets, such as ships, the multiple scattering effects have a significant influence on the EM scattering properties. Thus, a rectangular wave beam-based geometrical optics and physical optics (GO/PO) method is introduced to calculate the EM scattering properties, which is more efficient than the traditional GO/PO. The GO/PO method is suitable to simulate SAR images in which the EM scattering properties of each pulse need to be calculated. With these methods, the SAR images of the target on the sea surface are simulated. Based on the comparison of the SAR images between a static and dynamic sea surface, it is found that the region corresponding to the target is blurred and the texture of the dynamic sea is blurrier. The impact of multiple scattering and sea wave motion on target recognition are also analyzed with the SAR images that were generated under different conditions. Some strong scattering points appear when multiple scattering effects are considered. It is also found that the texture of the SAR images, corresponding to the sea surface, changes with the synthetic aperture time.

show abstract

Ship Detection and Feature Visualization Analysis Based on Lightweight CNN in VH and VV Polarization Images

Cited by 21 publications

References 48 publications

Small-Sized Ship Detection Nearshore Based on Lightweight Active Learning Model with a Small Number of Labeled Data for SAR Imagery

Small-Sized Ship Detection Nearshore Based on Lightweight Active Learning Model with a Small Number of Labeled Data for SAR Imagery

Self-Matching CAM: A Novel Accurate Visual Explanation of CNNs for SAR Image Interpretation

Simulation of a Wideband Radar Echo of a Target on a Dynamic Sea Surface

Contact Info

Product

Resources

About