Answers are grounded in this paper and cite their sources.
This paper reports on some aspects of using mixed radar and visible sensors to generate stereo-pairs that are suitable for stereo-mapping planimetric and elevation features. By choosing compromises to reduce geometric and radiometric disparities, the brain is able to perceive depth and to combine the complementary aspects of the radiometries from a VIR/SAR stereo-pair. Quantitative results of mapping and feature extraction from opposite and same-side ERS-SAR and SPOT-P stereo images are presented. In planimetry, accuracy of about 30m is obtained. Such accuracy is dependent on the stereo-pair, the stereo viewability, the image content and the feature definition. In altimetry, 20-m accuracies are obtained for both stereo-pairs. The strongest stereo geometry does not explicitly lead to better results, because with these mixed SAR/VIR stereo-pairs the major contributor to the elevation parallax is the SAR geometry. Finally, from the stereo-fusion of these two ERS-SAR/SPOT-P pairs, the radiometry of the SPOT-P images mainly contributes to the determination of the planimetric features with the quality of its image content; while the geometry of the ERS-SAR image mainly contributes to the determination of the elevation with its high sensitivity to the terrain relief.
No takes yet. Share an insight, caveat, or question.
Th Toutin (2000) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: