STUDY OF EXTRATION OF PLANIMETRIC CARTOGRAPHIC INFORMATION BY DIRECT LINEAR TRANSFORMATION WITH SINGULAR IMAGE OBTAINED BY REMOTELY PILOTED AIRCRAFT

Authors

DOI:

https://doi.org/10.21680/2447-3359.2022v8n1ID25675

Abstract

Traditionally in Photogrammetry, the extraction of three-dimensional cartographic data involves the use of two images at least. With a singular image, it is possible to obtain only planimetric information, however, external altimetry data are necessary for the execution of the process. The modeling of these information is done rigorously by the Collinearity Equations (CLN) which assumes the use of camera calibration parameters and the treatment of systematic errors. Alternatively, generalized transformations such as Direct Linear Transform (DLT) can be used, simplifying the process. An investigation was developed aiming to verify the performance of the DLT and CLN models in extracting cartographic information using a single image. The image was obtained by a low-cost camera attached to a remotely piloted aircraft. Maximum discrepancy values ??around 0.80 m were obtained for planimetric values for both models. The RMSE for planimetric discrepancies was 0.53 m by CLN, and 0.56 m by DLT. The cartographic quality (PEC-PCD) was classified in Class A by CLN and B by DLT, on a scale 1: 10000. DLT showed 90% efficiency in relation to rigorous formulation and operational advantages, in addition to not requiring camera parameters, which increases the possibility of using low-cost equipment.

Downloads

Download data is not yet available.

Published

14-06-2022

How to Cite

EURIQUES, J. F.; MITISHITA, E. A.; RAZOTO, V. G. .; KRUEGER, C. P.; VEIGA, L. A. K. STUDY OF EXTRATION OF PLANIMETRIC CARTOGRAPHIC INFORMATION BY DIRECT LINEAR TRANSFORMATION WITH SINGULAR IMAGE OBTAINED BY REMOTELY PILOTED AIRCRAFT . Notheast Geoscience Journal, [S. l.], v. 8, n. 1, p. 198–216, 2022. DOI: 10.21680/2447-3359.2022v8n1ID25675. Disponível em: https://periodicos.ufrn.br/revistadoregne/article/view/25675. Acesso em: 22 nov. 2024.

Issue

Section

Artigos