Xingkun He

Speckle suppression is a crucial preliminary step for synthetic aperture radar (SAR) image processing. Supervised despeckling approaches trained on synthetic datasets usually perform poorly in practice due to the unavailability of clean SAR images. Besides, the spatial correlation of speckle is rarely considered in many methods based on the fully developed speckle assumption. In this paper, we propose an unsupervised despeckling method to address these issues by combining online speckle generation and unpaired training. The method consists of two branches: the stop-gradient branch and the unpaired branch. First, the stop-gradient branch learns to generate the spatially correlated speckle. Then, the unpaired branch combines the generated speckle with the unpaired optical image to form pairs of training data for network parameter updates. More specifically, in order to generate the more realistic speckle in the stop-gradient branch, we design a speckle correction module with three SAR speckle priors: the threshold prior, the unit mean prior, and the correlation prior coupled with the weighted patch-shuffle. In the unpaired training, a hybrid loss function is employed, which takes spatial smoothness and detail protection into consideration. Afterwards, we combine the stop-gradient branch with the unpaired branch by the Siamese network to achieve alternate optimization of speckle generation and speckle removal. Finally, the optimization process in our method is analyzed theoretically. Qualitative and quantitative experiments demonstrate that the proposed method is comparable to the supervised despeckling approaches on synthetic datasets and outperforms several state-of-the-art unsupervised methods on real SAR datasets.

The frequency modulated continuous wave (FMCW) radar sensors have attracted much attention in short-range applications. However, when detecting a stationary target, an FMCW radar sensor obtains the time-varying beat signals due to its nonlinearity, which degrades the imaging quality. For this reason, a correction method for the nonlinearity of FMCW radar sensors is proposed based on the multisynchrosqueezing transform (MSST) in this paper. First, a reference beat signal is processed by the short-time Fourier transform (STFT) to generate the blurry time-frequency representation. Second, an iterative reassignment procedure is employed to concentrate the blurry time-frequency representation by the MSST method. Then, the concentrated time-frequency representation is used to estimate the nonlinear coefficients by ridge detection and polynomial fitting. Finally, the nonlinear beat signals are compensated using the nonlinear coefficients after time resampling. The effectiveness of the proposed method is validated by simulations and experiments. The results show that the proposed method can obtain more concentrated time-frequency representation and more accurate estimation of nonlinear coefficients, resulting in higher quality of range profiles and synthetic aperture radar (SAR) images. In addition, the proposed method has better noise robustness for the strongly time-varying and multicomponent beat signals.