Search

Extracting buildings from True Digital Ortho Maps often encounters occlusions and misidentifications, making it challenging to obtain complete, regular, and accurate building contours. To address this issue, we developed a building recognition process based on the Segment Anything Model, and proposed a novel regularization algorithm for building contour inference and fitting, which quantifies the confidence levels of contour points to accurately fit building contours from data containing substantial noise, and reformulates the fitting problem as progressive node classification tasks consisting of contour simplification, iterative regularization, and rationality assessment. In experimental evaluations, the proposed contour fitting algorithm achieved 97.99 % Intersection over Union (IoU), 95.39 % consistency with the standard contour edge count, and 88.06 % of cases with Hausdorff distances less than or equal to 15 pixels (30 cm), significantly outperforming comparative methods. Notably, it was the only contour regularization algorithm that improved IoU (1.03 %) compared to the original contours. The experimental results demonstrate that the proposed algorithm effectively suppresses noise and infers incomplete building contours, producing accurate and regular contours comparable to manual delineation. It is particularly suitable for buildings with near-orthogonal structures, exhibiting significant practical applicability and generalization potential.

Geometric structural information of objects in synthetic aperture radar (SAR) images is of great significance in target detection and analysis. The existing orthorectification methods of SAR images generally follow the traditional routes of optical images, resulting in suffering from the influence of layover, and bring difficulties in image analysis or target extraction without sufficient prior knowledge. Therefore, this article proposes a highest-projection-principle-based SAR true digital ortho map (TDOM) production method for the restoration of geometric structural information of targets. First, ground control information extracted from the high-precision digital surface model is applied to calibrate the geolocation error of original SAR images. Second, the highest projection principle and the visibility detection are utilized to reduce the influence of layover. To preserve the structural information of targets, multiview SAR TDOMs are integrated and compensated to generate a fused SAR TDOM image similar to optical view. Two Gaofen-3 (GF-3) SAR spotlight images are included to verify the performance of our proposed method. Experimental results are compared qualitatively in visual effects and quantitatively in geolocation accuracy. The results indicate that the produced SAR TDOMs can preserve the structural information of targets and provide more similar geometric structural information as optical images, which will effectively improve the efficiency in the interpretation of orthorectified SAR images.