9 January 2023 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Recently, convolutional neural network based methods have been studied for ship detection in optical remote sensing images. However, it is challenging to apply them to microwave synthetic aperture radar (SAR) images. First, most of the regions in the inshore scene include scattered spots and noises, which dramatically interfere with ship detection. Besides, SAR ship images contain ship targets of different sizes, especially small ships with dense distribution. Unfortunately, small ships have fewer distinguishing features making it difficult to be detected. In this article, we propose a novel SAR ship detection network called feature enhanced pyramid and shallow feature reconstruction network (FEPS-Net) to solve the above problems. We design a feature enhancement pyramid, which includes a spatial enhancement module to enhance spatial position information and suppress background noise, and the feature alignment module to solve the problem of feature misalignment during feature fusion. Additionally, to solve the problem of small ship detection in SAR ship images, we design a shallow feature reconstruction module to extract semantic information from small ships. The effectiveness of the proposed network for SAR ship detection is demonstrated by experiments on two publicly available datasets: SAR ship detection dataset and high-resolution SAR images dataset. The experimental results show that the proposed FEPS-Net has advantages in SAR ship detection over the current state-of-the-art methods.
6 March 2023 in IEEE Geoscience and Remote Sensing Letters
Most synthetic aperture radar (SAR) ship detectors based on convolutional neural networks (CNNs) needed preset anchor boxes to object classification and bounding box coordinate regression. However, the sparsity and unbalanced distribution of ships in SAR images mean that most anchor boxes are redundant. Thus, the anchor settings directly affect the performance and generalization ability of the detector. In addition, a variety in ship scales and the substantial interference of inshore backgrounds bring significant challenges to the SAR ship detector’s performance improvement. In this letter, a novel anchor-free-based detector, named feature balance and united attention (FBUA-Net), is proposed. We adopt a keypoint-based strategy to predict bounding boxes to eliminate the influence of anchors. Besides, we propose a global context-guided feature balanced pyramid (GC-FBP), which balances the semantic information at different levels of the feature pyramid by aggregation and averaging and uses the global context module (GCM) to learn global contextual information to construct long-range dependencies between ship targets and the background. Considering the interference of scattering noise to the detector, a united attention module (UAM) is designed to reduce the interference of surrounding noise by focusing on the spatial shape and scale size of ship targets in both the spatial and scale domains. Experimental results on the SAR ship detection dataset (SSDD) and high-resolution SAR images dataset (HRSID) datasets show that our detector achieves state-of-the-art (SOTA) performance. The source code can be found at https://github.com/so-bright/FBUA-Net .
23 June 2022 in IEEE Geoscience and Remote Sensing Letters
With the development of deep learning, remote sensing image (RSI) semantic segmentation has produced significant advances. The majority of existing methods use fully convolutional network (FCN) that lacks fine-grained multiscale representation and fails to extract global context information. Thus, we improve FCN by adding two modules—multiscale attention (MSA) and nonlocal filter (NLF). The MSA module enhances the network’s fine-grained multiscale representation capability and allows modeling the interdependencies of feature maps among different channels. The NLF module can capture global context information by sequential using fast Fourier transform (FFT), parameter learnable filters, and inverse FFT. By using MSA module for encoder and NLF module for decoder in the FCN framework, MSA and NLF network (MsanlfNet) can obtain both fine-grained multiscale spatial feature and global context information, thus achieving a balance between performance and computational effort. Experimental results on the remote sensing semantic segmentation public datasets demonstrate that our method can achieve better performance. The code is available at https://github.com/xyuanLin/MsanlfNet .
12 April 2022 in Scientific Reports
To alleviate the social contradiction between limited medical resources and increasing medical needs, the medical image-assisted diagnosis based on deep learning has become the research focus in Wise Information Technology of med. Most of the existing medical segmentation models based on Convolution or Transformer have achieved relatively sound effects. However, the Convolution-based model with a limited receptive field cannot establish long-distance dependencies between features as the Network deepens. The Transformer-based model produces large computation overhead and cannot generalize the bias of local features and perceive the position feature of medical images, which are essential in medical image segmentation. To address those issues, we present Triple Gate MultiLayer Perceptron U-Net (TGMLP U-Net), a medical image segmentation model based on MLP, in which we design the Triple Gate MultiLayer Perceptron (TGMLP), composed of three parts. Firstly, considering encoding the position information of features, we propose the Triple MLP module based on MultiLayer Perceptron in this model. It uses linear projection to encode features from the high, wide, and channel dimensions, enabling the model to capture the long-distance dependence of features along the spatial dimension and the precise position information of features in three dimensions with less computational overhead. Then, we design the Local Priors and Global Perceptron module. The Global Perceptron divides the feature map into different partitions and conducts correlation modelling for each partition to establish the global dependency between partitions. The Local Priors uses multi-scale Convolution with high local feature extraction ability to explore further the relationship of context feature information within the structure. At last, we suggest a Gate-controlled Mechanism to effectively solves the problem that the dependence of position embeddings between Patches and within Patches in medical images cannot be well learned due to the relatively small number of samples in medical images segmentation data. Experimental results indicate that the proposed model outperforms other state-of-the-art models in most evaluation indicators, demonstrating its excellent performance in segmenting medical images.