GEOBIA Vegetation Mapping in Great Smoky Mountains National Park with Spectral and Non-spectral Ancillary Information

Incorporating texture and topographic variables with spectral data from scanned color infrared aerial photographs significantly improved the accuracy of GEOBIA vegetation classification in Great Smoky Mountains National Park, increasing overall accuracy by 2.8 percent and Kappa by 5.0 percent. A separate study using multispectral Ikonos imagery found that elevation, aspect, slope, and proximity to streams also enhanced classification accuracy, achieving 76.6 percent overall accuracy and a Kappa of 0.57, closely resembling manual interpretations. These findings suggest that ancillary information can substantially aid GEOBIA vegetation mapping in rugged mountain areas, offering potential for more precise updates to vegetation inventories.

Kim, M., Madden, M., & Xu, B. (2010). GEOBIA Vegetation Mapping in Great Smoky Mountains National Park with Spectral and Non-spectral Ancillary Information. Photogrammetric Engineering & Remote Sensing, 76(2), 137–149. https://doi.org/10.14358/pers.76.2.137