In this paper, our approach is high-quality instance segmentation contains object detection in Remote Sensing imagery. In instance segmentation cross-entropy used as a loss function and intersection-over-union (IoU) used as a network performance measurement metric while in object detection, intersection over union (IoU) is often used to describe pos-itive/negative thresholds. Using IoU as a loss function can solve the problem between the loss function and the metric of the evaluation. We proposed a max-batch soft IoU training approach that eliminates the fixed IoU loss. randomness of the initial max-batch gradient descent (GD) technique. It resolves the IoU loss function's instability. However, our proposed method Cascade Mask R-CNN with max-batch soft IoU produces better results on the NWPU VHR-10 dataset for object detection and instance segmentation.

Ship detection in synthetic aperture radar (SAR) images is becoming a research hotspot. In recent years, as the rise of artificial intelligence, deep learning has almost dominated SAR ship detection community for its higher accuracy, faster speed, less human intervention, etc. However, today, there is still a lack of a reliable deep learning SAR ship detection dataset that can meet the practical migration application of ship detection in large-scene space-borne SAR images. Thus, to solve this problem, this paper releases a Large-Scale SAR Ship Detection Dataset-v1.0 (LS-SSDD-v1.0) from Sentinel-1, for small ship detection under large-scale backgrounds. LS-SSDD-v1.0 contains 15 large-scale SAR images whose ground truths are correctly labeled by SAR experts by drawing support from the Automatic Identification System (AIS) and Google Earth. To facilitate network training, the large-scale images are directly cut into 9000 sub-images without bells and whistles, providing convenience for subsequent detection result presentation in large-scale SAR images. Notably, LS-SSDD-v1.0 has five advantages: (1) large-scale backgrounds, (2) small ship detection, (3) abundant pure backgrounds, (4) fully automatic detection flow, and (5) numerous and standardized research baselines. Last but not least, combined with the advantage of abundant pure backgrounds, we also propose a Pure Background Hybrid Training mechanism (PBHT-mechanism) to suppress false alarms of land in large-scale SAR images. Experimental results of ablation study can verify the effectiveness of the PBHT-mechanism. LS-SSDD-v1.0 can inspire related scholars to make extensive research into SAR ship detection methods with engineering application value, which is conducive to the progress of SAR intelligent interpretation technology.

Ship detection in high-resolution synthetic aperture radar (SAR) imagery is a challenging problem in the case of complex environments, especially inshore and offshore scenes. Nowadays, the existing methods of SAR ship detection mainly use low-resolution representations obtained by classification networks or recover high-resolution representations from low-resolution representations in SAR images. As the representation learning is characterized by low resolution and the huge loss of resolution makes it difficult to obtain accurate prediction results in spatial accuracy; therefore, these networks are not suitable to ship detection of region-level. In this paper, a novel ship detection method based on a high-resolution ship detection network (HR-SDNet) for high-resolution SAR imagery is proposed. The HR-SDNet adopts a novel high-resolution feature pyramid network (HRFPN) to take full advantage of the feature maps of high-resolution and low-resolution convolutions for SAR image ship detection. In this scheme, the HRFPN connects high-to-low resolution subnetworks in parallel and can maintain high resolution. Next, the Soft Non-Maximum Suppression (Soft-NMS) is used to improve the performance of the NMS, thereby improving the detection performance of the dense ships. Then, we introduce the Microsoft Common Objects in Context (COCO) evaluation metrics, which provides not only the higher quality evaluation metrics average precision (AP) for more accurate bounding box regression, but also the evaluation metrics for small, medium and large targets, so as to precisely evaluate the detection performance of our method. Finally, the experimental results on the SAR ship detection dataset (SSDD) and TerraSAR-X high-resolution images reveal that (1) our approach based on the HRFPN has superior detection performance for both inshore and offshore scenes of the high-resolution SAR imagery, which achieves nearly 4.3% performance gains compared to feature pyramid network (FPN) in inshore scenes, thus proving its effectiveness; (2) compared with the existing algorithms, our approach is more accurate and robust for ship detection of high-resolution SAR imagery, especially inshore and offshore scenes; (3) with the Soft-NMS algorithm, our network performs better, which achieves nearly 1% performance gains in terms of AP; (4) the COCO evaluation metrics are effective for SAR image ship detection; (5) the displayed thresholds within a certain range have a significant impact on the robustness of ship detectors.

For the existence of speckles, many standard optical image processing methods, such as classification, segmentation, and registration, are restricted to synthetic aperture radar (SAR) images. In this work, an end-to-end deep multi-scale recurrent network (MSR-net) for SAR image despeckling is proposed. The multi-scale recurrent and weights sharing strategies are introduced to increase network capacity without multiplying the number of weights parameters. A convolutional long short-term memory (convLSTM) unit is embedded to capture useful information and helps with despeckling across scales. Meanwhile, the sub-pixel unit is utilized to improve the network efficiency. Besides, two criteria, edge feature keep ratio (EFKR) and feature point keep ratio (FPKR), are proposed to evaluate the performance of despeckling capacity for SAR, which can assess the retention ability of the despeckling algorithm to edge and feature information more effectively. Experimental results show that our proposed network can remove speckle noise while preserving the edge and texture information of images with low computational costs, especially in the low signal noise ratio scenarios. The peak signal to noise ratio (PSNR) of MSR-net can outperform traditional despeckling methods SAR-BM3D (Block-Matching and 3D filtering) by more than 2 dB for the simulated image. Furthermore, the adaptability of optical image processing methods to real SAR images can be enhanced after despeckling.

Object detection in synthetic aperture radar (SAR) imagery is a fundamental and challenging problem in the field of SAR imagery analysis for many fields like military, intelligence, commercial applications, etc. Object detection based on faster-regions convolutional neural network (Faster R-CNN) has lesser running time as compared to a convolutional neural network (CNN) in the detection process. Nowadays, anchor boxes are widely used in the detection model. This paper aims to provide an anchor box optimization method to improve ship detection accuracy in SAR imagery. By using Residual Network (ResNet-50) as a backbone in Faster R-CNN and it's compatible anchor sets better mean Average Precision (mAP) achieved. We compared the mAP with the two sets of anchor parameters, we found that in the ship detection process mAP achieves more than 4.29 % significant improvement.

Object detection in remote sensing imagery is a fundamental and challenging problem for accurate object detection. Faster-regions convolutional neural network (Faster R-CNN) in remote sensing imagery has reduced the running time of the detection networks. But in the traditional method of convolutional neural network (CNN), the process is very slow it takes too much time. Nowadays, anchor boxes are widely adopted in the stage of model detection. This paper aims to provide a method to improve the detection accuracy in remote sensing imagery by using Faster R-CNN and its compatible Residual Network (ResNet). We compared the mean Average Precision (mAP) with the two backbone network ResNet-50 (Reference Backbone) and ResNet-101 (Proposed Backbone) with the same set of anchor ratio, we found that mAP achieves more than 1.376% significant improvement.

In this paper, an array of 7 × 7 I - shaped resonators is used as Near-zero Index Metamaterial (NZIM) superstrate over Four Edges Gap-coupled Microstrip Antenna (FEGCOMA) for the purpose of gain enhancement. Prototypes of FEGCOMA and NZIM superstrate are fabricated to validate the results obtained in simulation. Measured results confirm that the FEGCOMA with the NZIM superstrate operates in S-band with a bandwidth is 16.06% centered at 2.49 GHz and exhibits a gain of 4.55 dBi to 6.75 dBi in the operating range, which is around 2 dBi more than the average gain of reference FEGCOMA.