Correcting bias in the rational polynomial coefficients of satellite imagery using thin-plate smoothing splines

An important and difficult point in the application of satellite imagery is refining the positioning model and improving the geometric accuracy. In this study, we focus on improvement in geometric accuracy and develop a new rational function model (RFM) refinement method. First, we derive the conversion relationship between the rigorous sensor model and the RFM, based on which we illustrate the approximate meaning of the zero-order and first-order terms of the rational polynomial coefficients (RPCs). Second, the correlation problem between RPCs and the influence of individual RPCs on geometric positioning accuracy are analyzed and verified. The dominant coefficients that determine geolocation are then identified. Finally, a new RFM refinement method based on direct correction of the dominant coefficients is proposed and validated. The experiments, conducted with ZY3-02 satellite imagery, indicate that the proposed method can effectively improve the geometric accuracy of satellite images.

Download Full-text

NUMERICAL EVALUATION AND APPLICATION-ORIENTED ANALYSIS FOR FORWARD AND INVERSE RATIONAL FUNCTION MODELS OF TERRAIN-INDEPENDENT CASE IN SATELLITE IMAGERY

Geodesy and Cartography ◽

10.3846/20296991.2014.962731 ◽

2014 ◽

Vol 40 (3) ◽

pp. 99-109 ◽

Cited By ~ 2

Author(s):

Sorosh Jahandideh ◽

Ali Azizi ◽

Nasser Najibi

Keyword(s):

Data Processing ◽

Satellite Imagery ◽

Control Points ◽

Virtual Control ◽

Polynomial Coefficients ◽

Rational Polynomial ◽

Aerospace Companies ◽

Independent Case ◽

Function Models ◽

And Control

Terrain-independent Rational Polynomial Coefficients (RPCs) are considered as most important part of the optical satellite and aerial imagery data processing especially those ones with high resolution since the proposed RPCs by the aerospace companies have some limitations in particular for using directly by the geoscientists in environmental studies and other Earth observation applications. While the inverse RPCs have more advantageous rather than direct ones, in this study, a new approach is presented in order to provide the inverse RPCs from direct ones and also to satisfy satellite imagery products users. In order to do this, first a spatial 3D-cubic is going to be fitted to the study area approximately including necessary altimetry layers numbers. Next, a range of virtual control points are being created in those altimetry layers randomly and then these points are going to be shifted to the image space by means of given direct RPCs. Hence, the inverse RPCs computes from the direct ones by space resection technique. Finally, the ground coordinates for the corresponding points have derived from different space intersection methodologies, direct RPCs and also inverse ones. Moreover, comparative tests have been developed to assess the effects of different altimetry layers numbers and also the number of virtual control points on the quality of derived inverse RPCs. It is demonstrated here that the precision of derived RPCs are increasing as much as the number of altimetry layers and control points increase. The proposed methodology, computations, data processing and results evaluation are discussed in details.

Download Full-text