Automatic radiometric normalization of multitemporal satellite imagery

Abstract

The linear scale invariance of the multivariate alteration detection (MAD) transformation is used to obtain invariant pixels for automatic relative radiometric normalization of time series of multispectral data. Normalization by means of ordinary least squares regression method is compared with normalization using orthogonal regression. The procedure is applied to Landsat TM images over Nevada, Landsat ETM+ images over Morocco, and SPOT HRV images over Kenya. Results from this new automatic, combined MAD/orthogonal regression method, based on statistical analysis of test pixels not used in the actual normalization, compare favorably with results from normalization from manually obtained time-invariant features.

Reburn severity in managed and unmanaged vegetation in a large wildfire

“…The two prefire images were then converted to reflectance and atmospherically corrected by using the COST method (34). We then used an automated ordination algorithm called ''multivariate alteration detection'' (35,36) that statistically located pseudoinvariant pixels, which were subsequently used in a reduced major axis regression to radiometrically normalize postfire to prefire images. We selected October imagery, despite the low sun angle in autumn, because we wanted to pair our imagery to the dates of historical aerial photos to aid an assessment of accuracy and because we wanted to capture the fire's effects without the confounding influence of spring green-up that may have occurred had we used imagery from the following summer.…”

Forest Monitoring Using Landsat Time Series Data: A Review

“…Pseudoinvariant features are spatially well-defined objects that are presumed to have stable reflectance properties over the period of the time series. A variety of methods have been developed for the selection of PIFs (e.g., Hall et al 1991;Schott et al 1998;Canty et al 2004;Paolini et al 2006). These approaches range from manual selection of features to cover the range of bright, midrange, and dark data values to automatic selection of invariant features using statistical methods such as multivariate alteration detection (MAD).…”

Implementación y evaluación del modelo Landsat Ecosystem Disturbance Adaptive Processing System (LEDAPS): estudio de caso en los Andes colombianos

“…Autores como Canty et al, 2004;Cheng et al, 2004;Lu et al, 2002;Paolini et al, 2006;Richter, 1998, tratan en sus artículos las dificultades en la implementación de modelos de corrección radiométrica y atmosférica en zonas de alta montaña para desarrollar estudios multitemporales. Sin embargo, en este estudio se evaluó la potencialidad de la implementación del modelo LEDAPS, como forma de corrección en reflectividad superficial en una zona de montaña dentro de los Andes colombianos, con cubierta de bosque tropical.…”

“…Teniendo en cuenta la variación anual que tiene la interacción Sol-Sensor-Superficie, se recomienda normalizar los datos (Canty et al, 2004) en estudios multitemporales; la corrección atmosférica por sí sola no garantiza el ajuste de la serie de tiempo al 100%. Además se ha encontrado que un importante aporte en la diferencia de los datos en el tiempo está relacionada a la interacción con el relieve y a la falta de ajuste en geometría dentro del dato entregado por el distribuidor.…”