3D Reconstrution with Rational Function Model
- PDF / 1,392,234 Bytes
- 9 Pages / 612 x 792 pts (letter) Page_size
- 91 Downloads / 177 Views
27
Photonirvachak
J. Indian Soc. Remote Sens. (March 2008) 36:27–35
RESEARCH ARTICLE
3D Reconstrution with Rational Function Model V. Nagasubramanian . P. V. Radhadevi . R. Ramachandran . R. Krishnan
Received: 20 April 2007 / Accepted: 12 November 2007
Keywords
Rational function model · Sensor model · Photo-grammetric processing · Image restitution
Abstract Rational Function Model (RFM) is the alternate sensor Model to the rigorous sensor model that allows end user to perform sensor-independent photogrammetric processing. Nowadays, commercial off-the-shelf (COTS) digital photogrammetric work stations have incorporated RFM as a method for image restitution. It is technically applicable to all types of airborne and space borne sensors. In this paper, we describe the derivations of the algorithmic
procedure for third order inverse and forward RFM method for 3-D reconstruction. Model accuracy is evaluated for aerial image, TK-350 Russian image and IRS-1C PAN image. The results ensure that properly constructed RFM are accurate enough to be used in place of the original rigorous models. The test results are reported and summarised.
Introduction
V. Nagasubramanian () . P. V. Radhadevi . R. Ramachandran . R. Krishnan Advanced Data Processing Research Institute Department of Space, Government of India, Hyderabad- 500 009, India
e-mail: [email protected]
During the last four to five years, the photogrammetric community has become aware of the use of RFM for image restitution. Dowmann and Dolloff, (2000) used the term ‘replacement sensor model’ as the generic term which uses ratios of polynomial functions to define the transformation from object space to image space. RFM model has been universally accepted and validated as an alternative sensor orientation model for high resolution satellite imagery like IKONOS, QUICK BIRD and SPOT-5 HRS. A number of papers (Tao and Hu, 2000; Hu and Tao,
28
J. Indian Soc. Remote Sens. (March 2008) 36:27–35
2002; Fraser and Hanley, 2003; Lutes, 2004; Grodecki et al., 2004; Chen et al., 2006) have been published on RFM indicating that high accuracy can be achieved. The use of satellite images within the mapping sector has increased tremendously. With the frequent launch of high-resolution satellites into the orbit, it has become mandatory that the processing and product generation system has to be sensor-independent. The satellite agencies may not like to release the complex camera model and meta data to the users. RFM is the answer for this. Some of the satellite agencies like Image Sat International (for EROS) give a grid (containing a set of image coordinates and corresponding ground coordinates) along with the image data from which Rational Function Co-efficients (RFCs) can be generated. Some other vendors like Space Imaging (for IKONOS) offer the RFCs straight away. If the camera parameters, ephemeris and attitude data are provided along with the image, we can generate a grid (using rigorous sensor model in the pre-processing) for fitting the RFCs. If the images along with
Data Loading...
