Self-supervised monocular depth estimation methods severely compromise accuracy in dynamic objects due to their static scene assumption. Existing approaches for dynamic scenes suffer from two critical shortcomings: 1) reliance on supervised segmentation models (requiring costly annotations) or computationally intensive multi-branch models to isolate moving objects, and 2) simple integration of 2D/3D motion flow without reliable supervision for dynamic objects. We propose AdaDepth, a two‑stage framework that jointly performs unsupervised scene decomposition and dynamic-aware depth learning. In the initial structural stage, our geometry-motion joint scene decomposition (GMoDecomp) module ensures the robust generation of a depth prior and simultaneously partitions the scene into multiple regions through the fusion of geometric and motion cues. In the region-adaptive refinement stage, we exploit the depth prior and decomposed regions to introduce motion-aware and geometry-consistent constraints, effectively improving depth estimation in dynamic scenes. AdaDepth achieves accurate depth prediction in highly dynamic scenes without relying on external labels or specialized segmentation models. Extensive experiments on KITTI, Cityscapes, and Waymo Open demonstrate its superiority over state-of-the-art approaches.

Swells significantly influence wave field energy distribution, coastal stability, ocean-atmosphere interface, and errors in ocean waves numerical models. High-frequency radar (HFR) turns out to be an effective device for swell observation. The traditional swell inversion method derived from classical equations describing the interaction of electromagnetic and oceanic waves is strongly confined by the inherently weak nature of the second-order radar signal. This paper proposes a novel deep learning-based approach named First-order Parallel Temporal-aware Neural Network (FPTNN) for swell waveheight inversion from HFR, which demonstrates a substantial improvement over the traditional method. The FPTNN method utilizes the robust first-order echo spectrum for swell inversion, differing from the previously used weak second-order spectrum. Compared with the traditional method, the FPTNN increases the effective output data percentage for swell waveheight from 6.90% to 80%, reduces the root mean square deviation (RMSD) from 0.35 m to 0.16 m, and lowers the mean absolute percentage deviation (MAPD) from 44.37% to 25.23%. Local wind field data can be incorporated as auxiliary information to refine the network's training process. In example data, the wind field information reduces the RMSD from 0.16 m to 0.14 m, and lowers the MAPD from 25.23% to 21.81%. Similar results were obtained at the HFR operating frequency of 8 MHz. The new approach significantly enhances the stability and utilization efficiency of radar echo spectra and reduces about half the error of inversion results. This advancement can be employed to promote the capability of HFR for routine swell measurement.

The accurate identification of components in the range-Doppler (RD) spectrum image is deemed crucial for radar interference suppression and parameter inversion in high-frequency hybrid sky-surface wave radar (HFSSWR) data processing. Previous studies have primarily focused on the manual selection of components in the RD spectrum image using regular boxes, with limited ability to simultaneously identify each component automatically. To address this issue, a small dataset comprising 364 images and 1190 annotations using real radar returns was constructed. Building upon the U-Net architecture, the lightweight aligned U-Net (LAU-Net) structure is introduced, where ResNet50 serves as the encoder, an alignment module is integrated to standardize image sizes, and focal loss is employed to mitigate the effects of class imbalance within the dataset on model performance. The test results demonstrate that the LAU-Net model outperforms the U-Net model, achieving improvements in frames per second (FPS) by 77.10%, mean pixel accuracy (mPA) by 13.48%, and mean IoU (mIoU) by 10.92%.

The high-frequency hybrid sky-surface wave radar (HSSWR) utilizes the ionosphere–ocean hybrid propagation channel and can detect targets over the horizon while preserving high detection precision. Owing to the uneven and ever-changing electron density distribution in the ionosphere, the phase path of electromagnetic waves propagating through it fluctuates irregularly, resulting in the widening, even splitting of the spectrum, and ultimately affecting the capacity of radar for detecting slow-moving targets. In this article, a novel parametric estimation and compensation method that combines the cubic phase function (CPF) and the high-order ambiguity function (HAF), referred to as the hybrid CPF-HAF method, is adopted to solve the problem of phase path contamination of the HSSWR signals. The direct wave is selected as the calibration signal and modeled by piecewise polynomial phase signals (PPSs). The maximum likelihood (ML) principle is adopted for selecting the PPSs’ orders, and the hybrid CPF-HAF method is employed to estimate the parameters of PPSs and reconstruct the correction signal. It is ultimately employed to compensate for the ionosphere-induced phase path contamination. Simulation results indicate that the CPF-HAF outperforms the HAF in terms of PPSs’ coefficients estimating accuracy. The processing results of measured data further demonstrate the efficiency of the proposed decontamination algorithm. After phase compensation, the dilated spectra of the sea clutter were sharpened, and the peak amplitudes significantly increased.

Abstract This paper proposes an image tampering detection algorithm based on sample guidance and individual camera device's convolutional neural network (CNN) features (SGICD‐CF) to address the challenges in the authenticity and integrity of images. Due to the development of the digital image processing technology, which makes image editing and processing, image tampering and forgery easy and lot simplified, thus solving the problem of image tamper detection, to maintain information security. The principle of SGICD‐CF assumes that pixels of the pristine image come from a single camera device, but on the contrary, if an image to be tested is spliced by multiple images from different cameras, then the pixels from the multiple camera devices will be detected. SGICD‐CF divides the image to be tested into 64 × 64 pixel image patches, extracts the camera‐related features and some camera model‐related information of image patches by source camera identification network (SCI‐Net) which is proposed by us, and obtains the classification confidence degree of the image patch. Furthermore, it determines whether the image patch contains foreign pixels according to the obtained confidence degree and finally determines whether the image was tampered according to the classification results of all the image patches, thus locating the tampered area. However, the experimental results show that SGICD‐CF can detect and locate the tampered area of an image accurately and our methods have a better performance than other existing methods. Our algorithm can achieve an average correct rate of 0.855 on the synthetic data set based on Dresden, which is higher than other existing detection methods.

High Frequency (HF) ground wave radar is widely used in maritime target detection and marine environment observation. Its performance is often degraded by co-frequency interference of the HF band. It is necessary to detect the environmental radio frequency for a HF radar to select the proper operation frequency. In this paper, a radio frequency detecting module is designed and equipped on a double-frequency all-digital multi-input and multi-output (MIMO) HF radar. The operational frequency bands are scanned in the operation of the radar to help select an unoccupied frequency in the band. The application of this module reduced the influence of RF interference on the HF radar. The improvement in the signal quality is obvious.

The detection performance of high-frequency surface wave radar (HFSWR) is highly susceptible to radio frequency interference (RFI). The fact that external RFIs will turn into a chirp signal and the fractional Fourier transform (FRFT) is a potent tool to analyze the chirp signal, motivates us to use FRFT in interference mitigation. In this letter, two new RFI suppression schemes are proposed, of which one is based on hybrid used of FRFT and autoregressive (AR) linear prediction algorithm, another is based on joint FRFT and recursive least squares (RLS) adaptive filtering algorithm. The simulation result indicates that the proposed schemes can effectively mitigate external RFIs even when the interferences cover Bragg peaks. The experimental result shows that the proposed schemes can effectively suppress RFI without losing echoes, and the signal to noise-plus-interference ratio is improved by 6 dB on average after RFI suppression.

Recently, high-frequency surface wave radar (HFSWR) has been widely applied in ocean surface dynamic parameter measurement. However, the radar echoes backscattered from ocean surface tend to be contaminated by the external radio frequency interference (RFI), and the mitigation of RFI becomes an intractable problem, especially for wide beam HFSWR. The HFSWR measured data are multichannel. The higher-order singular value decomposition (HOSVD) is an effective method to improve the accuracy of subspace estimation by exploiting this multidimensional structure. In this article, we develop an RFI mitigation method based on the HOSVD algorithm and orthogonal subspace projection. Simulations indicate that, compared with the previous orthogonal subspace projection RFI cancellation schemes, the proposed method has significant advantages in keeping the desired signals while suppressing the interference. The proposed method is applied to the experimental data of the HFSWR with severe RFI. The ocean surface currents inverted from the RFI-mitigated data agree reasonably with the tidal features in the radar detection area. After RFI mitigation, the performance of the HFSWR system is significantly improved on the effective current detection range and precision.

The 2016 Meranti typhoon was observed with two land-based high frequency radars when it passed through the Taiwan Strait before landing. The HF radar works at 7.9MHz with Bragg wave frequency of 0.29 Hz higher than typical spectrum peak frequency of typhoon generated wave. Sea surface wave field was retrieved using the cosine-2s directional function. By analyzing the curl of the wave direction vector, rotation center was obtained. Results shows that rotation center of measured wave field always appears at the left region of best typhoon path released by CMA.

The high-frequency hybrid sky-surface wave radar (HF HSSWR) has recently been used to monitor large-area sea states. However, most of the HF HSSWR detection methods are based on the assumption of a no-tilt and constant height ionospheric model, and the influences caused by uneven electron density are ignored. This paper proposes a new surface current inversion scheme for the HF distributed HSSWR system, which considers the unknown ionospheric state as a black box and extracts the key parameters to compute the surface current based on a scattering model. The computational formula of the component of the current vector is explored using spatial scattering theory instead of an approximate bistatic model. In addition, the Fourier series expansion method is applied to the HF data to extract the real first-order Bragg frequency. Subsequently, the grazing angle and the bistatic angle can be found by inversion using the first-order Bragg frequency formula after searching out the common scattering patch of two receiving stations. Simultaneously, the coordinate registration of the currents can also be determined. The feasibility and effectiveness of this new algorithm are verified with field experimental results by comparing the current vectors derived from HSSWR and traditional HF SWR. The RMS differences of the magnitude and direction of the current vectors within the core common area of the two detection systems are about 10.2 cm/s and 9.5°, respectively.

The conventional method of extracting ocean surface currents by high-frequency over-the-horizon radar is based on the fixed first-order Bragg frequency formula and ignores the effects caused by the environment, especially in near-shore areas. In this letter, a current inversion model based on 2-D Fourier series expansion was developed. The first-order Bragg frequency and the Doppler offset induced by radial current are dealt with the bivariate functions of group distance and azimuth angle in the proposed method. By solving an overdetermined matrix equation with the least-square fitting method, the current at each detection grid can be estimated. As the Bragg frequency obtained by this new algorithm is adaptive to the environment, the accuracy of current measurement will be improved. The feasibility and effectiveness of the new method are verified with simulations and experimental results. The currents estimated by the traditional method and the new algorithm are compared with two in situ buoys. Results indicate that the new algorithm possesses comparable accuracy for far-shore areas and better accuracy for near-shore areas when compared with the conventional method.

We consider estimating the direction-of-arrival (DOA) in the presence of sensor array error. In the proposed method, a blind signal separation method, the joint approximation and diagonalization of eigenmatrices algorithm, is implemented to separate the signal vector and the mixing matrix consisting of the array manifold matrix and the sensor array error matrix. Based on a new mixing matrix and the reconstruction of the array output vector of each individual signal, we propose a novel DOA estimation and sensor array error calibration procedure. This method is independent of array phase errors and performs well against difference of SNR of signals. Numerical simulations verify the effectiveness of the proposed method.

The spreading factor is considered as a key parameter that controls the concentration of the directional distribution of the wave energy. It has been confirmed by many scholars that there is a certain relationship between spreading factor and sea surface wind. In the application of high frequency surface wave radar (HFSWR), spreading factor is extracted from the ratio (RB) of power spectrum density (PSD) of positive (PB+) and negative (PB-) Bragg peaks. To extract accurate spreading factor, the premise is that the PSD of detection unit is as little as possible affected by the adjacent detection units. For narrow-beam radar, digital beamforming (DBF) is easy to meet requirements. But for broad-beam radar, it is very difficult. In this paper, a new scheme is proposed to extract spreading factor from broad-beam HFSWR data with the MUSIC-APES algorithm. Different from spatial filtering by DBF, MUSIC-APES directly estimates the azimuth of positive or negative Bragg waves and their echo amplitudes. For broad-beam radar, this scheme can still achieve high azimuth resolution and accurate amplitude estimation at the same time. It solves the biggest obstacle to extract the spreading factor from broad-beam HFSWR data. To verify the feasibility of this scheme, simulations and experiments are carried out to compare with DBF. The extraction accuracy is improved greatly. The results are very surprising. It shows that spreading factor and wind speed are highly relevant. This may be a new way to extract wind speed in the application of HFSWR.

One of the significant challenges in physical oceanography is getting an adequate space/time description of the ocean surface currents. One possible solution is the maximum cross-correlation (MCC) method that we apply to hourly ocean color images from the Geostationary Ocean Color Imager (GOCI) over five years. Since GOCI provided a large number of image pairs, we introduce a new MCC search strategy to improve the computational efficiency of the MCC method saving 95% of the processing time. We also use an MCC current merging method to increase the total spatial coverage of the currents, proving a 25% increase. Five-year mean and seasonal time-average flows are computed to capture the major currents in the area of interest. The mean flows investigate the Kuroshio path, support the triple-branch pattern of the Tsushima Warm Current (TC), and reveal the origin of the TC. The evolution of a warm core ring shed by the Kuroshio near the northeast coast of Honshu, Japan, is clearly depicted by a sequence of three monthly MCC composites. We capture the evolution of the Kuroshio meander over seasonal, monthly, and weekly time scales. Three successive weekly MCC composite maps demonstrate how a large anticyclonic eddy, to the south of the Kuroshio meander, influences its formation and evolution in time and space. The unique ability to view short space/time scale changes in these strong current systems is a major benefit of the application of the MCC method to the high spatial resolution and rapid refresh GOCI data.

A new algorithm using array beamforming for estimating waves and currents from marine X-band radar image sequences is proposed. Array beamforming is used to extract the wave directional spectrum from the image sequences; the image intensity time series is considered as the signal received by the “antennas” in conventional array beamforming. To retrieve the current velocity, the following feature is utilized: the total power reaches a maximum when a suitable current is used in the beamforming calculation. Because the new algorithm is based on individual pixel points, as opposed to the full image in the conventional method, the coordinate transformation procedure can be avoided. Because of the filter capacity in beamforming, moving vessels and shadowing have little effect on the estimate results, and the empirical modulation transfer function is not required to amend the wave spectrum. The wave and current retrieval scheme is validated using a simulation and by comparing results with data from an in situ buoy.

It is difficult to acquire lateral velocity in the current radar applications. In order to obtain the velocity vector, two or more radar sites are needed. However, it is of great cost, and it is difficult to achieve real-time measurement. In this letter, a scheme is proposed to measure lateral velocity based on radio interferometry with which two receiving channels are adopted to obtain an interference signal. A formula describing the relationship between the phase varying slope of the interference signal and the lateral velocity whose direction is parallel to the tangent of the ellipse whose foci are the position of two receiving antennas was deduced. To verify the feasibility of measuring the lateral velocity with the proposed scheme, simulations of the echo signal of the linear frequency modulation radar are carried out; the result shows that the proposed scheme is very effective. An experimental result with high-frequency radar of Wuhan University further verifies the feasibility of this scheme to measure lateral velocity.

A novel algorithm is developed to estimate the signal-to-noise ratio (SNR) for the significant wave height (SWH) inversion of the ocean wave fields imaged by X-band radar. The proposed algorithm utilizes the geometric relationship of the linear dispersion relation for gravity sea surface waves to separate signal and noise without the determination of sea surface current. Compared with the traditional SNR-based method, the SWH derived by using this proposed method matches well with that measured by an in situ buoy nearby.

A method for extracting sea surface wind direction information from bistatic high-frequency (HF) radar Doppler spectra is presented. By analogy to the monostatic case, the ratio of the intensities of the positive and negative bistatic Bragg peaks is used to derive the (ambiguous) wind direction. For bistatic operation, the reference is taken with respect to the scattering ellipse normal rather than the radar beam direction. The method is shown to be valid based on simulated bistatic HF radar Doppler spectra. Wind direction is also extracted from the bistatic radar data collected on the Southern China coast. Comparison between the radar-measured wind directions and those obtained from the Advanced Scatterometer shows good agreement.