H addressed in this paper can be divided into 3 parts
H addressed within this paper might be divided into 3 parts: block adjustment of multi-slice photos, panoramic stitching image generation and RFM construction, and block adjustment assisted by high-resolution reference data. The planar residual of Information A in image space was 1.687527 pixels, and also the planar residual of three images in Data B was close to one particular pixel. These benefits demonstrate that we improved positioning accuracy and supplied an accuracy assure for the RFM construction. Adopting the piecewise affine transformation model to create the panoramic stitching image, the accuracies of stitching images have been all within 1 pixel, meeting the requirements of seamless stitching and achieving the stitching accuracy measured in [14]. Compared together with the object-space-oriented algorithm in [191], the difference of object space spatial positioning accuracy was within 0.three m, reaching constant positioning accuracy. five. Conclusions Within this paper, we proposed a brand new technique for multi-slice satellite images stitching and geometric model building. The image-space-oriented algorithm has no precise sensor (Z)-Semaxanib In Vivo parameters and attitude and needs orbit information. Users may perhaps locate it tough to acquire and method information, relying only on the original slice image information and facts and RFM. The panoramic stitching image achieves sub-pixel level stitching accuracy. Furthermore, the RFM positioning accuracy is consistent using the object-space-oriented algorithm, meeting the user’s subsequent Nitrocefin Formula application specifications. We employed RFM as the coordinate conversion model, which integrates the coordinate conversion relationship of image space and object space continuity. The proposed approach, in comparison with all the image-space-oriented algorithm, can establish a clear geometric object-image relationship. Also, as opposed to the object-space-oriented algorithm, the proposed method is simpler and can establish the geometric object-image connection of panoramic stitching images without the need of sensor parameters. In addition, the proposed method has speed advantages in image processing, becoming computationally efficient with a smaller sized quantity. Nonetheless, our proposed approach is determined by image matching. This means that the image-stitching and RFM positioning accuracy is usually affected if we lack the image texture options. Moreover, the RFM in the original slice image is generated after calibrating the geometric sensor distortion and its mounting angle error, platform stability, as well as other parameters. ThisRemote Sens. 2021, 13,14 ofmethod requires high positioning accuracy to make sure the geometric top quality in the stitching image.Author Contributions: Conceptualization, L.W., Y.Z. (Yan Zhang) and T.W.; Data curation, L.L.; Formal evaluation, Z.Z. and Y.Y.; Methodology, L.W., Y.Z. (Yan Zhang) and T.W.; Computer software, L.W.; Validation, Y.Y.; Visualization, Y.Z. (Yongsheng Zhang) and Z.Z.; Writing–Original draft, L.W. and Y.Z. (Yan Zhang); Writing–Review and Editing, T.W. and Y.Z (Yongsheng Zhang). All authors have read and agreed towards the published version of your manuscript. Funding: This research received no external funding. Institutional Critique Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest.
Academic Editor: John J. Qu Received: 16 September 2021 Accepted: 16 November 2021 Published: 20 NovemberPublisher’s Note: MDPI stays neutral with regard to jurisdic.
