|
|
|
A Technique for Assessing the Accuracy of Sub-Pixel Impervious Surface Estimates Derived from Landsat Tm Imagery
 |
|
| Abstract: | We developed a technique for assessing the accuracy of sub-pixel derived estimates of impervious surface extracted from LANDSAT TM imagery. We utilized spatially coincident
sub-pixel derived impervious surface estimates, high-resolution planimetric GIS data, vector--to-
raster conversion -methods and raster GIS overlay methods to derives a level of agreement
watershed, a small (14 km2) sub-watershed in the mid-Atlantic physiographic region. From the planimetric data we produced a per-pixel truth estimate of impervious surface percent as a means for from LANDSAT TM imagery- The spatial technique allows for multiple accuracy assessment approaches. Results showed that even though per-pixel based estimates of the accuracy of the sub-pixel data were poor (Error Matrix Overall Accuracy = 28.41, Kappa coefficient of agreement = 0. 185), the accuracy of the impervious surface percentage estimated using whole- area and rank correlation approaches were much more accurate (Relative Percent Correct =70.85, Spearman correlation = 0.608). Our findings suggest that per-pixel based approaches to the accuracy assessment of sub-pixel classified data need to be approached with some caution. Per-pixel based approaches may underestimate the actual whole-area accuracy of the material-of- interest map, as derived from sub-pixel methods, when applied over large geographic areas raster overlay technique easily extracted the data necessary to derive these assessments. The suited for the assessment of data at the sub-pixel level, it can be utilized - or altered - to derive the accuracy of any classified dataset where higher resolution digital truth data is available.
|
|
| Citation: | Jarnagin, S. T., D. B. Jennings, D. W. Ebert, R. S. Lunetta, and J. G. Lyon. A Technique for Assessing the Accuracy of Sub-Pixel Impervious Surface Estimates Derived from Landsat Tm Imagery.Ross Lunetta & John Lyon (ed.), Remote Sensing and GIS Accuracy Assessment, Chapter19. Taylor & Francis Books, Inc., Boca Raton, FL, (2004). |
|
|
| Contact: |
Chris Siebert - (702) 798-2234 or siebert.christopher@epa.gov
|
|
| Division: |
Environmental Sciences Division |
|
| Branch: |
Landscape Ecology Branch |
|
| Product Type: |
Book Chaptr |
|
| Published: |
07/27/2004 |
|
|
|
