Engineering topics

Leberl, F. W.

Publications and source records attributed to Leberl, F. W..

6 records

NASA NTRS1991

Measurement of surface microtopography

Acquisition of ground truth data for use in microwave interaction modeling requires measurement of surface roughness sampled at intervals comparable to a fraction of the microwave wavelength and extensive enough to adequately represent the statistics of a surface unit. Sub-centimetric measurement accuracy is thus required over large areas, and existing techniques are usually inadequate. A technique is discussed for acquiring the necessary photogrammetric data using twin film cameras mounted on a helicopter. In an attempt to eliminate tedious data reduction, an automated technique was applied to the helicopter photographs, and results were compared to those produced by conventional stereogrammetry. Derived root-mean-square (RMS) roughness for the same stereo-pair was 7.5 cm for the automated technique versus 6.5 cm for the manual method. The principal source of error is probably due to vegetation in the scene, which affects the automated technique but is ignored by a human operator.

Wall, S. D.↗

NASA NTRS1989

Measurement of surface microtopography using helicopter-mounted stereo film cameras and two stereo matching techniques

A common problem in acquiring ground-truth data for use in microwave interaction modeling is the capture of surface roughness data that are both sampled at distances comparable to a fraction of the wavelength and extensive enough to represent the surface statistics in at least one resolution cell of the microwave remote sensor used. A technique has been developed for acquiring the necessary photogrammetric data using twin 70-mm film cameras mounted on a helicopter boom. The apparatus is described, and the accuracy with which ground surface roughness can be characterized using this device is estimated. Both standard and cross-correlation methods were used for data reduction. Stereogrammetry is compared with a completely automated image-matching technique. Dense disparity images were generated from the helicopter stereo pairs. Using interior orientation parameters supplied by the camera manufacturers, and assuming that exterior orientation parameters remain constant between control target and test field photography, an extremely dense DEM (digital elevation model) for a test field has been derived. Results are compared, and accuracy estimates are presented.

Wall, S. D.↗

NASA NTRS1986

Radar stereomapping techniques and application to SIR-B images of Mt. Shasta

One of the goals of the SIR-B experiment was the definition of optimum radar incidence and intersection angles for radargrammetric stereoscopy by comparing the results from three separate data-reduction methods. To this end three overlapping images of the prime radargrammetric site (Mt. Shasta in northern California) were obtained, fewer than anticipated. This paper describes the mathematical basis for radar stereomapping, along with preliminary results from one of the methods (using a so-called 'analytical' stereoplotter) for the Mt. Shasta site. Height coordinate errors range from 60 to 170 m depending on the density (number per ground area) of ground control points used.

Leberl, F. W.↗

NASA NTRS1985

On stereo viewing of SAR images

Stereo viewing is indispensable for the measurement of homologue points in two overlapping images and the subsequent reconstruction of the three-dimensional topographic relief. Aspects of stereo viewing geometry are discussed along with problems of radar stereoscopy, taking into account stereo evaluation with an exaggeration factor, the viewability of real images, and radar stereoscopic computations. Attention is given to the definition of radar stereoscopic exaggeration, questions of image simulation, and the accuracy obtained with parallax measurements in radar images. An evaluation with a large set of about 40 radar stereo models demonstrates that same-side arrangements provide good stereo viewability. This result was confirmed for aircraft radar with look angles off-nadir of 60-80 deg and intersection angles between 0.2 deg and 23 deg, and for satellite radar (Seasat) with look angles of 20 deg and interaction angles of 1.2 deg - 4.8 deg.

Leberl, F. W.↗

NASA NTRS1980

Application of imaging radar to mapping

The present state of radar mapping is reviewed. Emphasis is on radargrammetric mapping with single images, stereo pairs, and block adjustment. Applications to thematic mapping are also addressed. Examples presented concern radar mosaicking, sea ice study and extraterrestrial mapping.

Leberl, F. W.↗

NASA NTRS1979

Accuracy aspects of stereo side-looking radar

The geometry of the radar stereo model and factors affecting visual radar stereo perception are reviewed. Limits to the vertical exaggeration factor of stereo radar are defined. Radar stereo model accuracies are analyzed with respect to coordinate errors caused by errors of radar sensor position and of range, and with respect to errors of coordinate differences, i.e., cross-track distances and height differences.

Leberl, F. W.↗