Ship detection on synthetic aperture radar (SAR) imagery has many valuable applications for both civil and military fields and has received extraordinary attention in recent years. The traditional detection methods are insensitive to multiscale ships and usually time‐consuming, results in low detection accuracy and limitation for real‐time processing. To balance the accuracy and speed, an end‐to‐end ship detection method for complex inshore and offshore scenes based on deep convolutional neural networks (CNNs) is proposed in this paper. First, the SAR images are divided into different grids, and the anchor boxes are predefined based on the responsible grids for dense ship prediction. Then, Darknet‐53 with residual units is adopted as a backbone to extract features, and a top‐down pyramid structure is added for multiscale feature fusion with concatenation. By this means, abundant hierarchical features containing both spatial and semantic information are extracted. Meanwhile, the strategies such as soft non‐maximum suppression (Soft‐NMS), mix‐up and mosaic data augmentation, multiscale training, and hybrid optimization are used for performance enhancement. Besides, the model is trained from scratch to avoid learning objective bias of pretraining. The proposed one‐stage method adopts end‐to‐end inference by a single network, so the detection speed can be guaranteed due to the concise paradigm. Extensive experiments are performed on the public SAR ship detection dataset (SSDD), and the results show that the method can detect both inshore and offshore ships with higher accuracy than other mainstream methods, yielding the accuracy with an average of 95.52%, and the detection speed is quite fast with about 72 frames per second (FPS). The actual Sentinel‐1 and Gaofen‐3 data are utilized for verification, and the detection results also show the effectiveness and robustness of the method.

The spaceborne interferometric synthetic aperture radar (InSAR) is expected to measure the sea surface height (SSH) with high accuracy over a wide swath. Since centimeter-level accuracy is required to monitor the ocean sub-mesoscale dynamics, the high accuracy implies that the altimetric errors should be totally understood and strictly controlled. However, for the dynamic waves, they move randomly all the time, and this will lead to significant altimetric errors. This study proposes an analytical method for the dynamic wave-related errors of InSAR SSH measurement based on the wave spectrum and electromagnetic scattering model. Additionally, the mechanisms of the dynamic wave-related errors of InSAR altimetry are analyzed, and the detailed numerical model is derived. The proposed analytical method is validated with NASA’s Surface Water and Ocean Topography (SWOT) project error budget, and the Root-Mean-Square Errors (RMSEs) are in good agreement (0.2486 and 0.2470 cm on a 0.5 km² grid, respectively). Instead of analysis for a typical project, the proposed method can be applied to different radar parameters under multiple sea states. The RMSEs of Ka-band under low sea state, moderate sea state, and high sea state are 0.2670, 1.3154, and 6.6361 cm, respectively. Moreover, the RMSEs of X-band and Ku-band are also simulated and presented. The experimental results demonstrate that the dynamic wave-related errors of InSAR altimetry are not sensitive to the frequencies but are sensitive to the sea states. The error compensation method is necessary for moderate and higher sea states for centimetric accuracy requirements. This can provide feasible suggestions on system design and error budget for the future interferometric wide-swath altimeter.

In this paper, a soft threshold method based on a wavelet transform, and a decomposition reconstruction method, are proposed to preprocess photoplethysmograph (PPG) signals. First, the soft threshold method is applied to remove the high-frequency noise. The wavelet decomposition and reconstruction method are then used to adjust the baseline drift in the PPG signals. The aim is to obtain a higher signal-to-noise ratio and smoother PPG signals. Finally, the validity of the proposed signal preprocessing method is verified using the general indicators of RMSE, the smoothness, and the starting point variance.