Faced with severe operating conditions, rolling bearings tend to be one of the most vulnerable components in mechanical systems. Due to the requirements of economic efficiency and reliability, effective fault diagnosis methods for rolling bearings have long been a hot research topic of rotary machinery fields. However, traditional methods such as support vector machine (SVM) and backpropagation neural network (BP‐NN) which are composed of shallow structures trap into a dilemma when further improving their accuracies. Aiming to overcome shortcomings of shallow structures, a novel hierarchical algorithm based on stacked LSTM (long short‐term memory) is proposed in this text. Without any preprocessing operation or manual feature extraction, the proposed method constructs a framework of end‐to‐end fault diagnosis system for rolling bearings. Beneficial from the memorize‐forget mechanism of LSTM, features inherent in raw temporal signals are extracted hierarchically and automatically by stacking LSTM. A series of experiments demonstrate that the proposed model can not only achieve up to 99% accuracy but also outperform some state‐of‐the‐art intelligent fault diagnosis methods.

Due to the advantages of Synthetic Aperture Radar (SAR), the study of Automatic Target Recognition (ATR) has become a hot topic. Deep learning, especially in the case of a Convolutional Neural Network (CNN), works in an end-to-end way and has powerful feature-extracting abilities. Thus, researchers in SAR ATR also seek solutions from deep learning. We review the related algorithms with regard to SAR ATR in this paper. We firstly introduce the commonly used datasets and the evaluation metrics. Then, we introduce the algorithms before deep learning. They are template-matching-, machine-learning- and model-based methods. After that, we introduce mainly the SAR ATR methods in the deep-learning era (after 2017); those methods are the core of the paper. The non-CNNs and CNNs, that is, those used in SAR ATR, are summarized at the beginning. We found that researchers tend to design specialized CNN for SAR ATR. Then, the methods to solve the problem raised by limited samples are reviewed. They are data augmentation, Generative Adversarial Networks (GAN), electromagnetic simulation, transfer learning, few-shot learning, semi-supervised learning, metric leaning and domain knowledge. After that, the imbalance problem, real-time recognition, polarimetric SAR, complex data and adversarial attack are also reviewed. The principles and problems of them are also introduced. Finally, the future directions are conducted. In this part, we point out that the dataset, CNN architecture designing, knowledge-driven, real-time recognition, explainable and adversarial attack should be considered in the future. This paper gives readers a quick overview of the current state of the field.

The presence of complex electromagnetic noise in the ocean significantly impacts the accuracy of ship shaft-rate electric field signal detection, necessitating the development of an effective denoising method to enhance detection precision. Nevertheless, traditional denoising methods encounter issues like low frequency resolution, challenging threshold configuration, and mode mixing. This study introduces a method that integrates variational mode decomposition (VMD) with multi-window spectral subtraction (MSS). The intrinsic mode functions (IMFs) of noisy signals are extracted using VMD, and the noise components within different IMFs are identified. The spectral features of both signal and noise within different IMFs are leveraged to eliminate noise signals via MSS. Subsequently, the denoised components of IMFs are rearranged to derive the denoised ship shaft-rate electric field signals, achieving noise reduction across various frequency bands. Following validation using simulation signals and empirical data, the noise reduction efficacy of VMD-MSS surpasses that of alternative methods, demonstrating robust performance even at low signal-to-noise ratios. The marine electromagnetic noise is effectively suppressed in the empirical data, while preserving the characteristics of ship’s shaft-rate signals, thereby validating the method’s efficacy and demonstrating its practical engineering value.

Deep learning-based object detectors have demonstrated remarkable performance in recent years, including their application in Synthetic Aperture Radar (SAR) ship detection. However, these detectors often adopt backbones borrowed from classification models, requiring pre-training as a preliminary step. In addition, these borrowed backbones have several limitations that hinder their effectiveness when applied to SAR images. To overcome these drawbacks and improve accuracy, we designed a lightweight backbone called SAR ship detection backbone and an attention frontend network called SAR ship detection frontend to train ship detectors from scratch. The former is a meticulously designed backbone that leverages techniques such as cross-spatial partial dense blocks to extract features and has improved efficiency in accurately locating small ships. The latter is a carefully designed frontend network equipped with an attention module for prediction. The attention module comprises a spatial attention path, channel attention path and identity mapping path. It facilitates the network's learning of more meaningful spatially and channel-wise features. In addition, we defined the scales and aspect ratios of bounding boxes in datasets. The one2one label assigning strategy was used to match every anchor box with the corresponding ground truth. Complete intersection over union loss was also used to improve the performance of regression. The results obtained from public datasets show a favorable tradeoff between accuracy and time consumption. We also found that the detectors trained from scratch relied on the volume of the training dataset. The findings of this work can be used as a reference for researchers to design SAR ship detectors tailored to their specific requirements and demands.

Although advanced deep learning techniques have significantly improved SAR ship detection, accurately detecting small ships remains challenging due to their limited size and the few appearance and geometric clues available. In order to solve this problem, we propose several small object detection strategies. The backbone network uses space-to-depth convolution to replace stride and pooling. It reduces information loss during down-sampling. The neck integrates multiple layers of features globally and injects global and local information into different levels. It avoids the inherent information loss of traditional feature pyramid networks and strengthens the information fusion ability without significantly increasing latency. The proposed intersection-of-union considers the center distance and scale of small ships specifically. It reduces the sensitivity of intersection-of-union to positional deviations of small ships, which is helpful for training toward small ships. During training, the smaller the localization loss of small ships, the greater their localization loss gains are. By this, the supervision of small ships is strengthened in the loss function, which can make the losses more biased toward small ships. A series of experiments are conducted on two commonly used datasets, SSDD and SAR-Ship-Dataset. The experimental results show that the proposed method can detect small ships successfully and thus improve the overall performance of detectors.

The Doppler information in complement to the intensity of radar backscatter has been verified useful for ocean SAR application. Accurate estimate of the ocean surface motions requires accurate estimate of the Doppler centroid. When we process the real SAR data, we find that different methods yield different mean values of the estimates. After a careful analysis, we find that it is caused by the underestimation of the centroids for conventional methods. The estimate bias depends on the Doppler centroid itself. To overcome this problem, we carefully study the conventional methods and the Cramér-Rao bound of the estimation. Based on the maximum-likelihood principle, a novel method is proposed. By adaptively adjusting the center and the bandwidth of the weighting function after a coarse estimation of the spectrum, the symmetry of the spectrum is maintained. The estimate bias of conventional methods is eliminated. It is a fully optimal method meeting the Cramér-Rao bound no matter whether the true Doppler centroid is high or low. Both the simulated and the real SAR data validate our novel method. In the simulation, an estimate error of tens of Hz could be avoided when the true Doppler center is hundreds of Hertz. In real SAR estimation, an accuracy of 3-6 Hz can be improved. The result of the estimation is consistent with the simulation and theoretical analysis, validating the efficiency of our improved method.