On-orbit geometric calibration is a key technology to guarantee the geometric quality of high-resolution optical satellite imagery. In this paper, we present an approach for the on-orbit geometric calibration of high-resolution optical satellite imagery, focusing on two core problems: constructing an on-orbit geometric calibration model and proposing a robust calculation method. First, a rigorous geometric imaging model is constructed based on the analysis of the major error sources. Second, we construct an on-orbit geometric calibration model through performing reasonable optimizing and parameter selection of the rigorous geometric imaging model. On this basis, the calibration parameters are partially calculated with a stepwise iterative method by dividing them into two groups: external and internal calibration parameters. Furthermore, to verify the effectiveness of the proposed calibration model and methodology, on-orbit geometric calibration experiments for ZY1-02C panchromatic camera and ZY-3 three-line array camera are conducted using the reference data of the Songshan calibration test site located in the Henan Province, China. The experimental results demonstrate a certain deviation of the on-orbit calibration result from the initial design values of the calibration parameters. Therefore, on-orbit geometric calibration is necessary for optical satellite imagery. On the other hand, by choosing multiple images, which cover different areas and are acquired at different points in time to verify their geometric accuracy before and after calibration, we find that after on-orbit geometric calibration, the geometric accuracy of these images without ground control points is significantly improved. Additionally, due to the effective elimination of the internal distortion of the camera, greater geometric accuracy was achieved with less ground control points than before calibration.

High-quality inner FoV (Field of View) stitching is currently a prerequisite step for photogrammetric processing and application of image data acquired by spaceborne TDI CCD cameras. After reviewing the technical development in the issue, we present an inner FoV stitching method based on sensor geometry and projection plane in object space, in which the geometric sensor model of spaceborne TDI CCD images is used to establish image point correspondence between the stitched image and the TDI CCD images, using an object-space projection plane as the intermediary. In this study, first, the rigorous geometric sensor model of the TDI CCD images is constructed. Second, principle and implementation of the stitching method are described. Third, panchromatic high-resolution (HR) images of ZY-1 02C satellite and triple linear-array images of ZY-3 satellite are utilized to validate the correctness and feasibility of the method. Fourth, the stitching precision and geometric quality of the generated stitched images are evaluated. All the stitched images reached the sub-pixel level in precision. In addition, the geometric models of the stitched images can be constructed with zero loss in geometric precision. Experimental results demonstrate the advantages of the method for having small image distortion when on-orbit geometric calibration of satellite sensors is available. Overall, the new method provide a novel solution for inner FoV stitching of spaceborne TDI CCD images, in which all the sub-images are projected to the object space based on the sensor geometry, performing indirect image geometric rectification along and across the target trajectory. At present, this method has been successfully applied in the daily processing system for ZY-1 02C and ZY-3 satellites.

To generate an orthoimage product, mosaicking is a necessary process, and seam-based mosaicking of orthoimages is popular. However, many of these methods only focus on the generation of seamlines between two adjacent orthoimages, so the final generated mosaicking image depends on the order of compositing. To address this shortcoming, this paper presents an initial seamline network generation method based on improved seeded region growing. The basis of this method is the use of raster data rather than vector calculation, which is used with the area Voronoi diagrams with overlap (AVDO)-based method. First, the effective area of each image and overlap regions between adjacent images are determined. Then, the improved seeded region growing algorithm obtains the seamlines of each overlap region. The main improvement is that the boundary lines of overlap regions, rather than individual points, are chosen as seeds of the seeded region growing algorithm. These seeds grow simultaneously until growing regions overlap. The generated separatrix of growing regions is regarded as the seamline in the overlap region. At the same time, the cut result of the image’s effective area is obtained. After that, these generated cut images are intersected to generate the effective mosaic polygon (EMP) of the image. Finally, all generated EMPs are vectorized to form the initial seamline network. In this way, the proposed method can process any kind of overlap region, and the final generated seamline network has no relation to the order of the image compositing. The experimental results demonstrate that the presented method is feasible and can achieve higher accuracy than the previous AVDO-based method.

Blending processing based on seamlines in image mosaicking is a procedure designed to obtain a smooth transition between images along seamlines and make seams invisible in the final mosaic. However, for high-resolution aerial orthoimages in urban areas, factors such as projection differences, moving objects, and radiometric differences in overlapping areas may result in ghosting and artifacts or visible shifts in the final mosaic. Such a mosaic is not a true reflection of the earth’s surface and may have a negative impact on image interpretation. Therefore, this paper presents a multi-resolution blending method considering changed regions to improve mosaic image quality. The method utilizes the region change rate (RCR) to distinguish changed regions from unchanged regions in overlapping areas. The RCR of each region is computed using image segmentation and change detection methods. Then, a mask image is generated considering changed regions, and Gaussian and Laplacian pyramids are constructed. Finally, a multi-resolution reconstruction is performed to obtain the final mosaic. Experimental results from digital aerial orthoimages in urban areas are provided to verify this method for blending processing based on seamlines in mosaicking. Comparisons with other methods further demonstrate the potential of the presented method, as shown in a detailed comparison in three typical cases of the seamline passing by buildings, the seamline passing through buildings, and the seamline passing through areas with large radiometric differences.