Recent deep learning (DL) detectors adopted by radar or sonar (RS) are normally trained with transfer learning, where the typical workflow is to pretrain a convolutional neural network (CNN) on external large-scale classification datasets (e.g., ImageNet) as the backbone and then finetune the entire detector on detection datasets. Though transfer learning could effectively avoid overfitting, transferred models are usually redundant and might not generalize well on RS datasets. To achieve high generalization and to eliminate the dependence on transfer learning, a self-trained target detection method is established by including Automatic Deep Learning (AutoDL) to design optimal detectors. This self-trained target detection consists of three stages. First, a derived classification dataset (DCD) consisting of image blocks of targets and backgrounds is derived from detection datasets. Then, a memory-efficient Differentiable Architecture Search algorithm with flexible search space and large inputs (FL-DARTS), which is characterized by its predefined multistride convolutions, poolings, and unique super-structure, is proposed to automatically design and self-train optimal CNNs on DCDs. Finally, self-trained AutoDL detectors are implemented with the automatic backbone designed by FL-DARTS. We evaluated three self-trained AutoDL detectors on the public SAR ship detection dataset (SSDD) and the self-made sonar common target detection dataset (SCTD). The experiments show that while the number of parameters of automatic backbones designed for SSDD and SCTD are only 11.8% and 15.2% of that of ResNet50, self-trained AutoDL detectors implemented with automatic backbones significantly outperform their transfer learning detectors and achieve state-of-the-art detection precisions and high detection speeds. Data, codes are publicly available.

In multi-receiver Synthetic Aperture Sonars (SAS), the traditional non-stop-hop-stop model normally ignores the intra-pulse motion of sonar platform. However, such intra-pulse motion brings the Doppler scale effect to the echo signal, which will lead to the change of Carrier Frequency (CF) and Frequency Modulation (FM) rate of echo signal and may eventually result in image defocus. In this paper, considering the intra-pulse motion, the time-varying time delay and range history of multi-receiver SAS were derived by introducing the range time into the range history, and then the Doppler scale signal model was established. After approximating the derived time-varying range history, the two-dimensional (2D) spectrum was derived by using the shear theorem. When compared with the traditional 2D spectrum, an interesting phenomenon was found that the change of FM rate has no effect on 2D spectrum. Actually, only two additional terms, i.e., range cell migration (ARCM) and azimuth modulation (AAM), were introduced into the new 2D spectrum due to the change of CF. Besides, the condition of neglecting the ARCM was given by further analyzing the 2D spectrum, while the AAM is usually small and can be easily compensated together with the ARCM correction. Finally, a Chirp-z transform (CZT) algorithm for the Doppler scale signal model was proposed to correct the ARCM, which can simultaneously correct the ARCM and compensate the AAM (although not required) in 2D frequency domain by phase multiplication. The effectiveness of the signal model and proposed algorithm is verified by simulations and field data.

Introduction: China has more than 300 million current smokers. There is a controversy over smokers’ risk of COVID-19 infection. Smoking is a risk factor for COVID-19 disease progression, and the outbreak of COVID-19 may change people’s smoking behaviors. This study assessed people’s...

The compensation of slant range motion errors needs to be carried out after range pulse compression in the signal processing of multiple-receiver synthetic aperture sonar (SAS). Whereas range pulse compression cannot be performed as the first step in multiple-receiver SAS chirp scaling (CS) algorithm, as it requires the transmitted signal to be linearly frequency modulated. For this reason, it is difficult to combine CS algorithm with motion compensation. Besides, those imaging algorithms based on interpolation, such as range-Doppler algorithm and wavenumber domain algorithm, have plenty of computation load and large phase errors. Aimed at these problems, a CZT algorithm for multiple-receiver SAS in the non-stop-hop-stop mode is proposed in this paper. The proposed algorithm can correct range cell migration utilizing chirp-z transform, which avoids interpolation. Furthermore, the algorithm is easily combined with motion compensation as it can process pulse-compressed data directly. This paper describes the algorithm in 9 key steps, including the principle of correcting range cell migration utilizing chirp-z transform in discrete domain. In the simulation and sea trial data processing, the proposed algorithm provides image quality equal to the CS algorithm, and the algorithm is slightly more efficient than CS algorithm when combined with motion compensation.

Precise orientation information facilitates target recognition of sonar images. Traditional orientation estimation methods, including the Hough transform method, have poor antinoise abilities, so they cannot achieve high estimation precision for sonar images, which usually have low signal-to-noise ratios. The convolutional neural network (CNN) regression method is sensitive to image affine transformations and requires extensive computation, so it cannot achieve strong robustness or high speed. To achieve high-precision, high-speed, and robust orientation estimation of sonar image targets, we present a novel orientation estimation method via the wavelet subimage energy ratio (WSER). The WSER varies with the rotation angles of images and has the highest value when the long axes of targets are vertical. It is translational and scale invariant and does not need supervised training. Therefore, we propose estimating orientations of sonar image targets by finding the max of the WSER to achieve high precision, high speed, and strong robustness to the translation and scale transformations of images. The results of experiments on a self-made sonar image dataset show that the mean absolute error (MAE) of the proposed method is 7.9°, while the MAEs of the CNN regression method and traditional methods are 10.1 and 22.9°, respectively. In addition, the proposed method is six times faster than the CNN regression method. The proposed orientation estimation method has also been applied to align the orientation of sonar images and CNN can achieve state-of-the-art performance for rotated target recognition using the aligned images. Data and codes are publicly available.

In synthetic aperture imaging, the interpulse Doppler shift is commonly used, while the intrapulse Doppler shift is ignored. However, for the linear frequency modulation signal, the intrapulse Doppler shift can cause an additional range cell migration (ARCM) and a phase error term. Additionally, the ARCM may exceed a range resolution cell in certain cases of multireceiver synthetic aperture sonar (SAS), which will result in image defocus. Therefore, the intrapulse Doppler shift in multireceiver SAS should not be ignored in such cases. In this study, the condition of the negligible intrapulse Doppler shift is presented, that is, the moving distance of the sonar platform within a pulse should be much less than an azimuth resolution cell. Based on the multireceiver accurate time-delay model, a multireceiver SAS intrapulse Doppler shift signal model is developed by introducing the intrapulse Doppler frequency into the echo signal. After simplifying the accurate range history under the narrow-beam assumption, the range-Doppler spectrum is derived. Finally, a multireceiver SAS chirp scaling algorithm considering the intrapulse Doppler shift (IDS-CSA) is derived. The proposed algorithm can correct the ARCM and compensate the phase error (although not required), and then obtain a well-focused synthetic aperture image. The effectiveness of the proposed model and algorithm is verified by simulations and field data. The results show that the proposed algorithm provides a better image quality than the conventional multireceiver CSA when the condition of the negligible intrapulse Doppler shift is not met.