Image registration is the basis for the joint interpretation of synthetic aperture radar (SAR) and optical images. However, the significant nonlinear radiation difference (NRD) and the geometric imaging model difference render the registration quite challenging. To solve this problem, both traditional and deep learning methods are used to extract structural information with dense descriptions of the images, but they ignore that structural information of the image pair is coupled and often process images separately. In this paper, a deep learning-based registration method with a co-attention matching module (CAMM) for SAR and optical images is proposed, which integrates structural feature maps of the image pair to extract keypoints of a single image. First, joint feature detection and description are carried out densely in both images, for which the features are robust to radiation and geometric variation. Then, a CAMM is used to integrate both images’ structural features and generate the final keypoint feature maps so that the extracted keypoints are more distinctive and repeatable, which is beneficial to global registration. Finally, considering the difference in the imaging mechanism between SAR and optical images, this paper proposes a new sampling strategy that selects positive samples from the ground-truth position’s neighborhood and augments negative samples by randomly sampling distractors in the corresponding image, which makes positive samples more accurate and negative samples more abundant. The experimental results show that the proposed method can significantly improve the accuracy of SAR–optical image registration. Compared to the existing conventional and deep learning methods, the proposed method yields a detector with better repeatability and a descriptor with stronger modality-invariant feature representation.

Compared with traditional mine detection methods, UAV-based measures are more suitable for the rapid detection of large areas of scatterable landmines, and a multispectral fusion strategy based on a deep learning model is proposed to facilitate mine detection. Using the UAV-borne multispectral cruise platform, we establish a multispectral dataset of scatterable mines, with mine-spreading areas of the ground vegetation considered. In order to achieve the robust detection of occluded landmines, first, we employ an active learning strategy to refine the labeling of the multispectral dataset. Then, we propose an image fusion architecture driven by detection, in which we use YOLOv5 for the detection part, to improve the detection performance instructively while enhancing the quality of the fused image. Specifically, a simple and lightweight fusion network is designed to sufficiently aggregate texture details and semantic information of the source images and obtain a higher fusion speed. Moreover, we leverage detection loss as well as a joint-training algorithm to allow the semantic information to dynamically flow back into the fusion network. Extensive qualitative and quantitative experiments demonstrate that the detection-driven fusion (DDF) that we propose can effectively increase the recall rate, especially for occluded landmines, and verify the feasibility of multispectral data through reasonable processing.

Geosynchronous (GEO) spaceborne–airborne very high-frequency ultra-wideband bistatic synthetic aperture radar (VHF UWB BiSAR) can conduct high-resolution and wide-swath imaging for ocean scenes. However, GEO spaceborne–airborne VHF UWB BiSAR imaging faces some challenges such as the geometric configuration, huge amount of echo data, serious range–azimuth coupling, large spatial variance, and complex motion error, which increases the difficulty of the high-efficiency and high-precision imaging. In this paper, we present an improved bistatic fast factorization backprojection (FFBP) algorithm for ocean scene imaging using the GEO satellite-unmanned aerial vehicle (GEO-UAV) VHF UWB BiSAR, which can solve the above issues with high efficiency and high precision. This method reconstructs the subimages in the orthogonal elliptical polar (OEP) coordinate system based on the GEO satellite and UAV trajectories as well as the location of the imaged scene, which can further reduce the computational burden. First, the imaging geometry and signal model of the GEO-UAV VHF UWB BiSAR are established, and the construction of the OEP coordinate system and the subaperture imaging method are proposed. Moreover, the Nyquist sampling requirements for the subimages in the OEP coordinate system are derived from the range error perspective, which can offer a near-optimum tradeoff between precision and efficiency. In addition, the superiority of the OEP coordinate system is analyzed, which demonstrates that the angular dimensional sampling rate of the subimages is significantly reduced. Finally, the implementation processes and computational burden of the proposed algorithm are provided, and the speed-up factor of the proposed FFBP algorithm compared with the BP algorithm is derived and discussed. Experimental results of ideal point targets and natural ocean scenes demonstrate the correctness and effectiveness of the proposed algorithm, which can achieve near-optimal imaging performance with a low computational burden.