André Rothkirch

Large-scale bidirectional reflectance model for urban areas

A bidirectional reflectance factor (BRF) model for urban areas for pixel sizes of more than 500 m/spl times/500 m is developed. Possible applications include albedo calculation, improvement of classification, and change detection algorithms, simulated global BRF maps, and refinement of atmospheric correction algorithms. The model combines the BRF effects at several scales (street grid, intermediate-sized objects, and microscale). The authors present modeling results as well as a comparison with measured data. The basic features of the urban BRF are the hotspot and the independence of its shape from wavelength in the range 450 nm/spl les//spl lambda//spl les/2300 nm. An analytical function that approximates the model is proposed for easy implementation and fast computation.

Modeling the directional reflectance (BRDF) of a corrugated roof and experimental verification

Remotely sensed images with a pixel size of about 1 m can nowadays be acquired by airborne scanners and in the near future also by high resolution satellites. With such a high spatial resolution, remotely sensed data of urban areas can resolve structures like a roof into the different surface segments with different inclinations, e.g. in the case of a gabled roof. The authors have measured the BRDF (bidirectional reflectance distribution function) effects thoroughly on a roof covered with corrugated (sinusoidally shaped) roof tiles and on a sample of flat roof tiles. They modeled the shape of the corrugated tiles by a cosine function and assumed that every infinitesimal surface patch of the roof tile has a BRDF proportional to the BRDF of the flat roof tile. Model results and measurements agree well. The most critical parameters are the ratio height over wavelength of the sinusoidal roof tiles and the intensity of the specular peak of the surface patch. It is possible to retrieve these parameters from the measurements.

Change detection with 1 m resolution satellite and aerial images

We propose an optimization of a computer based change detection technique based on Iterative Principal Component Analysis (IPCA). We determine and evaluate the changes between an airborne and a spaceborne multispectral image data set, the latter recorded by the commercial satellite IKONOS-2. The change detection algorithm proved to be applicable to large remotely sensed data sets. A vegetation filter, a shadow filter and an oversaturation filter improved the accuracy of the results. When applying all filters more than 80 percent of the objects with changes due to construction activity are detected by the IPCA algorithm. The false alarm rate (change of an object indicated but not verified) is about 5 percent.

Error analysis for BRDF measurements at the European goniometric facility

This study determines the measurement errors for laboratory bidirectional reflectance (BRDF) measurements at the European Goniometric Facility (EGO, Ispra, Italy). The average total error depends strongly on the spectral signature of the sample, varying from 4% to up to 14%. The spatial illumination heterogenity is identified as the largest error source. Its effects increase with increasing zenith angles. It is shown that the instrument noise of the detector SE590 (Spectron Engineering) is a function of the maximum signal in a measured spectrum. Based on this result, a procedure to define the wavelength range with acceptable measurement errors is presented. Different ways to calculate the irradiance are compared. It is shown that calculating a constant irradiance for each incidence angle is in better agreement with the reciprocity principle than assuming a different irradiance for each illumination and viewing geometry.

BRDF field studies for remote sensing of urban areas

The focus of this paper lies on two in situ measurement campaigns: First, the BRDF of man‐made surfaces is determined using a goniometer table. The BRDF of the samples is dominated by the specular peak. The data can be described by a simple analytical function. Second, the BRDF of a corrugated roof is measured. Results from the first campaign are used to model the results of the second campaign. It is found that the BRDF of the corrugated roof is very sensitive to the intensity of the specular peak and the shape (corrugation) of the roof tiles. It is possible to determine the shape parameter of the corrugated tiles from the data. Measurement techniques to eliminate skylight effects are presented.

Width of the specular peak perpendicular to the principal plane for rough surfaces

The bidirectional reflectance distribution function (BRDF) model developed by Torrance and Sparrow [J. Opt. Soc. Am. 57, 1105-1114 (1967)] is used to describe the specular reflection of rough surfaces. We compare this model with the BRDF measurements of four manmade surfaces with different roughnesses. The model can be used to describe the basic features of the measured BRDFs. We found that the width of the specular peak perpendicular to the principal plane decreases strongly with an increasing illumination zenith angle in the data as well as in the model. A model analysis shows that the width is approximately proportional to the cosine of the illumination angle theta(i), and the deviations are determined by the roughness of the surface. This relationship is accompanied by an increase in reflectance in the specular direction in the principal plane that is 1/cos theta(i) stronger than the increase for a perfectly smooth surface.

BRDF measurements on urban materials using laser light: Equipment characteristics and estimation of error sources

Multiangular measurements were made on urban surface materials (e.g., roof coverings) using the European Goniometer Facility (EGO) in Ispra, Italy. The intention is to analyze polarization effects in multiangular reflection data under controlled illumination conditions. In this study we determine the accuracy of the measurements. A HeNe laser at wavelength λ = 632.8 nm as illumination source and a spectroradiometer with a mounted polarizer as detector were used. We discuss different detector characteristics and determine the contributions of different error sources (for example noise or uncertainty of adjustments). The results of the calculation led to a relative accuracy of the reflectance values of ∼ 4.3% for the given setup. Measured reflectance values for 4 polarization states are presented for 3 samples (Spectralon, roof tile, and roof paper) in particular results describing the data of the roof tile by a model of Torrance and Sparrow. The model is able to describe the specular reflection for like‐polarization measurements. It describes the strong specular behavior at ss‐polarization as well as the much weaker specular reflection at pp‐polarization which are both shifted to large viewing zenith angles. Using different coefficients to describe the contribution from the diffuse component, relative deviations between measurements and model are about 11 % on average. Cross‐polarization measurements are approximately Lambertian with deviations of ∼ 5% on average.

Determination of the angular sensitivity of a multispectral line scanner from image data

The DAEDALUS AADS 1268 is a multispectral line scanner with 11 spectral channels and 716 pixels per line. This paper presents a post-flight calibration method to correct the data for dependence of the detector sensitivity on the scanning angle. An area has to be found where only negligible BRDF effects are expected across the principle plane for zenith angles smaller than the maximum scanning angle. The area does not need to be homogenous, but it must extend over a whole scan line. The runway of the Nuremberg airport was chosen. The pixels of the scan line acquired when crossing the runway at right angles were divided by the respective pixels of the perpendicular overflight after georegistration. The resulting angular sensitivity functions show variations up to 15% (depending on channel), similar to findings from a laboratory experiment three years earlier. A comparison with laboratory data from a DAEDALUS operated in Australia shows similar results, except for channels 2 and 8.