THE EFFECT OF ILLUMINATION ON STEREO DTM QUALITY: SIMULATIONS IN SUPPORT OF EUROPA EXPLORATION

• Published in: ISPRS annals of the photogrammetry, remote sensing and spatial information sciences remote sensing and spatial information sciences, 2016-06, Vol.III-4, p.103-110
• Main Authors: Kirk, R. L., Howington-Kraus, E., Hare, T. M., Jorda, L.
• Format: Article
• Language: English
• Subjects: Angles (geometry); Computer simulation; Illumination; Incidence angle; Matching; Pixels; Topography
• ISSN: 2194-9050; 2194-9042; 2194-9050
• DOI: 10.5194/isprs-annals-III-4-103-2016

We have investigated how the quality of stereoscopically measured topography degrades with varying illumination, in particular the ranges of incidence angles and illumination differences over which useful digital topographic models (DTMs) can be recovered. Our approach is to make high-fidelity simulated image pairs of known topography and compare DTMs from stereoanalysis of these images with the input data. Well-known rules of thumb for horizontal resolution (>3–5 pixels) and matching precision (~0.2–0.3 pixels) are generally confirmed, but the best achievable resolution at high incidence angles is ~15 pixels, probably as a result of smoothing internal to the matching algorithm. Single-pass stereo imaging of Europa is likely to yield DTMs of consistent (optimal) quality for all incidence angles ≤85°, and certainly for incidence angles between 40° and 85°. Simulations with pairs of images in which the illumination is not consistent support the utility of shadow tip distance (STD) as a measure of illumination difference, but also suggest new and simpler criteria for evaluating the suitability of stereopairs based on illumination geometry. Our study was motivated by the needs of a mission to Europa, but the approach and (to first order) the results described here are relevant to a wide range of planetary investigations.