Qiming Yuan

Sensor errors, platform ephemeris errors and auxiliary digital elevation model (DEM) errors can have an impact on the positioning accuracy of synthetic aperture radar (SAR) images. Utilizing corner reflectors for geometric calibration is a common way to improve parameter accuracy and therefore positioning accuracy. In this paper, we propose an energy-based geometric self-calibration method without relying on corner reflectors. Based on the radiometric and geometric properties of SAR images, the energy of SAR orthophoto over the rugged mountainous areas is taken as objective function. The conjugate gradient method is used to estimate fast time offset and slow time offset by maximizing the objective function, which achieves the equivalent compensation for positioning errors. SAR images from Radarsat-2, COSMO-SkyMed, TerraSAR-X, GaoFen-3, LuTan-1 and ChaoHu-1 satellites were used for the experiments, and the experimental results demonstrate the effectiveness and applicability of the proposed method. The accuracy evaluation results of corner reflectors and field-measured checkpoints show that the proposed method improves the positioning errors of SAR images from different satellites from tens of meters to less than ten meters. Our proposed method not only provides a new perspective on the geometric calibration of SAR, but also reduces the maintenance cost and the workload of external calibration during the daily operation of spaceborne SAR, which is of great significance for low-cost commercial satellites.

Water-body type classification plays a vital role in ecological conservation, water resource management, and urban planning. Accurate classification can aid decision-makers in understanding the functions of different water-body types, providing key information for urban planning and promoting harmony between human activities and the natural environment. Despite extensive research in the field of water-body segmentation, exploration in the water-body type classification community is not as widespread. Therefore, this article proposes a novel water-body type classification method based on a two-step deep learning model, decomposing water-body type classification into water-body segmentation and water-body type identification. Especially, this method constructs a unique data strategy by organically integrating backscatter features, polarimetric features, and DEM features, providing the model with rich and comprehensive information. In the first step, the segmentation network uses the fused feature to extract all water-body from Synthetic Aperture Radar (SAR) images. Subsequently, the extracted water-body are combined with the input data, forming a multi-feature input for the identification network to distinguish between natural and artificial water-body. During this process, a swarm intelligence optimization algorithm is employed to explore the optimal hyperparameters of the network, including those of the segmentation and identification networks. Finally, the proposed method is assessed using extensive experiments on water-body segmentation tasks, water-body type identification tasks, and joint water-body type classification tasks. This article not only provides a new perspective in the field of water-body type classification but also demonstrates the immense potential of deep learning network hyperparameter optimization and feature fusion in solving such problems.

In recent years, the application of deep learning for water-body detection in Synthetic Aperture Radar (SAR) images has seen extensive development. However, a significant proportion of these works primarily concentrate on enhancing and optimizing the model structure, with inadequate exploration of the potential impact of hyperparameter settings, a critical determinant of model performance. Thus, to fully exploit the power of deep learning in water-body detection from SAR images, this letter presents a diversified optimization strategy that revolves around the Dung Beetle Optimizer-Convolutional Neural Network (DBO-CNN) model, complemented by characteristic fusion and decision-level fusion. The DBO-CNN model employs DBO algorithm to search for optimal hyperparameter of CNN model for bolstering the performance of water-body detection in SAR images. To further enhance the performance, the DBO-CNN model uses unique input data which is constructed by integrating the polarimetric characteristic obtained from H/α and model-based polarization decomposition methods with backscatter characteristic. Finally, two decision-level fusion methods are proposed to optimize detection results, enhancing the recall and Intersection over Union (IoU) to 96.5% and 91.5%, respectively. In summary, Spaceborne SAR images, with the application of polarization decomposition and neural network, provides new insights and in-depth understanding for detecting water-body.

Synthetic Aperture Radar (SAR) is an active microwave sensor with all-day/night and all-weather detection capability, which is crucial for detecting surface water resources. Surface water-body such as rivers, lakes, reservoirs, and ponds usually appear as dark areas in SAR images. Accurate and automated extraction of these water bodies can provide valuable data for the management and strategic planning of surface water resources and effectively help prevent and control drought and flood disasters. However, most deep learning-based methods rely on manually labeled samples for model training and testing, which is inefficient and may introduce errors. To address this problem, this paper proposes a novel water-body detection method that combines optimization algorithms and deep learning techniques to automate water-body label extraction and improve the accuracy of water-body detection. First, this paper uses a swarm intelligence optimization algorithm, Dung Beetle Optimizer (DBO), to optimize the initial cluster center of the K-means clustering algorithm, which is called the DBO-K-means (DK) method. The DK method divides the training images into four categories and extracts the water bodies in them to generate the water-body labels required for deep learning model training and testing, and the whole process does not require human intervention. Then, the labels generated by DK and training data set images are fed into the Classifier–Optimizer (CO) for training. The classifier performs a dense classification task at the pixel level, resulting in an initial result image with blurred boundaries of the water body. Then, the optimizer takes this preliminary result image and the original SAR image as input, performs fine-grained optimization on the preliminary result, and finally generates a result image with a clear water-body boundary. Finally, we evaluated the accuracy of water-body detection using multiple performance indicators including ACC, precision, F1-Score, recall, and Kappa coefficient. The results show that the values of all indicators exceed 93%, which demonstrates the high accuracy and reliability of our proposed water-body detection method. Overall, this paper presents a novel DK-based approach that improves the automation and accuracy of deep learning methods for detecting water bodies in SAR images by enabling automatic sample extraction and optimization of deep learning models.