Premium
Rational function model for sensor orientation of IRS‐P6 LISS‐4 imagery
Author(s) -
Nagasubramanian V.,
Radhadevi P. V.,
Ramachandran R.,
Krishnan R.
Publication year - 2007
Publication title -
the photogrammetric record
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.638
H-Index - 51
eISSN - 1477-9730
pISSN - 0031-868X
DOI - 10.1111/j.1477-9730.2007.00447.x
Subject(s) - pixel , orientation (vector space) , rational function , computer science , photogrammetry , function (biology) , artificial intelligence , image (mathematics) , georeference , polynomial and rational function modeling , computer vision , digital elevation model , polynomial , data mining , mathematics , remote sensing , pattern recognition (psychology) , geography , geometry , mathematical analysis , evolutionary biology , physical geography , biology
This paper explores the application of a rational function model (RFM) as a replacement sensor model for IRS‐P6 LISS‐4 imagery. The rational polynomial coefficients (RPCs), initially generated using a rigorous sensor model (RSM) through direct georeferencing, are bias‐compensated with a minimum number of ground control points and are used for various photogrammetric applications such as digital elevation model and ortho‐image generation. The performance of RFM and RSM is compared in the sensor modelling of LISS‐4 imagery over long strips. Results show that accuracies achieved using RFM are within 1 pixel (worst case) of the accuracies derived using RSM. Error variation as a function of the number of quasi‐control points (anchor points) used for RFM fitting as well as model errors with respect to the length of the image strip are analysed. System‐level accuracy does not deteriorate when the RFM is fitted up to a length of 1200 km. Absolute positioning accuracy of 1·5 pixels (∼9 m) is achieved from bias‐compensated RPCs. The results demonstrate the potential of RFM as a replacement sensor model. This allows standardisation of product generation packages to handle multiple sensors.