IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · 2023 · 23 citations · 47 references
ComputingComputer ScienceOcean EngineeringComputer VisionMachine VisionRemote SensingDeep LearningImage Feature EngineeringRadar Clutter SuppressionConvolutionConvolutional Neural NetworksDeep Learning-based DetectionSpeckle InterferenceSpeckle Noise ReductionSynthetic Aperture RadarVisual SaliencyVisual Saliency Detection
Recently, deep learning-based methods for synthetic aperture radar (SAR) ship detection have made remarkable advancements. However, most existing methods primarily focus on achieving high detection accuracy by employing complex models, leading to an increase in computational costs. Additionally, some methods do not adequately consider the impact of speckle noise interference. To address these challenges, we propose a multiscale ship detector, called visual saliency-lightweight ship detector (VS-LSDet), utilizing visual saliency and lightweight convolutional neural network (CNN). First, a visual saliency enhancement module (VSEM) is proposed as a preprocessing step to visually highlight the ships and weaken the impact of speckle noise in the image. Second, a lightweight backbone called ghost-shuffle net (GSNet) is designed. We introduce two types of ghost-shuffle blocks (GSBlocks) as basic convolution blocks by introducing ghost convolution (GSConv) to reduce the model complexity, and channel shuffle operation to enhance the representation ability of the feature map. Then, we propose a multi-shape dilated convolution block (MDCB) incorporated into GSNet to enlarge its receptive fields, further improving the detector's performance. Finally, a hybrid attention module (HyAM) is proposed, it leverages both spatial and channel information within the feature map. HyAM can emphasize ship-related features while suppressing irrelevant features from the background in feature map. Experimental results on public SAR ship datasets demonstrate that, compared to other ship detection methods, VS-LSDet achieves higher detection accuracy with lower model complexity. Specifically, on the SSDD dataset, the AP value of VS-LSDet is 97.51%, with 2.53 M parameters and 6.21 GFLOPs.
47
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Ross Girshick, Jeff Donahue, Trevor Darrell et al. · 2014 · 31.8K citations
Kaiming He, Georgia Gkioxari, Piotr Dollár et al. · 2017 · 28.8K citations
Feature Pyramid Networks for Object Detection
Tsung-Yi Lin, Piotr Dollár, Ross Girshick et al. · 2017 · 28.7K citations
Squeeze-and-Excitation Networks