Document Type : Research Paper

Author

• Khosrou Khajeh (Translator)

Abstract

In order to achieve a precise method of converting PAN photography from a spatial image to a spatial object, a mathematical model has been designed in which at least one GCP (Ground Control Point) is used to determine the external justification of images. This mathematical model was originally prepared for Spot satellite imagery, which uses multicolinear equations to provide a satellite path model, while the satellite height difference would be formulized as a function of time using high-order polynomial modeling. Initial orbital information is obtained from the given data and refinement is also done using the iterative least squares solution. This mathematical model was tested for three different states: (1) a single image (2) strip (caused by a detector during orbit passage), and (3) a stereo pair. For states (1) and (2), a mean error of 9.1 m in width and 7.6 in length can be obtained by using a ground control point (GCP) for modeling. Using a ground control point was identified from a map of scale 1:50000. Accuracy was 38.3 meters in width, 42.6 meters in length, and 23.8 in height for the stereo pair. The results confirm the model and somewhat suggest ideas with which PAN (IRS-1C) will be able to influence the future developments of photogrammetry and cartography. The software mentioned in this paper for modeling orbit altitude is part of SOFTSPACE, a digital photogrammetric station for working with IRS-1C PAN and Spot imagery stereo data, a multifunction software package for preprocessing, photo conversion into map, a Digital Terrain Model (DTM) and features compilation modules.

Keywords

• IRS-1C PAN
• orbit state modelling
• single image
• stereo pair
• 3D Positioning Precision