Image analysis for facility siting: a comparison of low- and high-altitude image interpretability for land use/land cover mapping
H.M. Borella, U.S. Nuclear Regulatory Commission (NRC), California Univ., Santa Barbara, CA (USA), J. Estes, C. Ezra, Joseph Scepan, L. R. Tinney · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 1982
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · Office of Scientific and Technical Information · report · 1982
Abstract
This study documents for two test sites in pennsylvania the interpretability of commercially acquired low-altitude and existing high-altitude aerial photography in terms of time, costs, and accuracy for Anderson Level II land use/land cover mapping. Information extracted from the imagery is to be used in the evaluation process for siting energy facilities. Land use/land cover maps were drawn at 1:24,000 scale using commerically flown color infrared photography. Maps at 1:62,500 scale were made using existing high-altitude photography obtained from the Unied States Geological Surveys; EROS Data Center. The USGS-developed Anderson classification scheme (Level II) was used.
Detailed accuracy of the maps generated by manual image analysis was assessed by employing a stratified unaligned sampling procedure with additional samples selected and analyzed to insure adequate class representation. Both area weighted and by class accuracies were documented and field-verified. A discrepancy map was also drawn to illustrate differences in classifications between the two map scales. Map accuracy statistics as well as figures on production costs and the time needed to compile the two map sets were compiled and tabulated.
Results show that the 1:24,000 scale map set was more accurate (99% to 94% area weighted) than the 1:62,500 scale set, especially when sampled by class (96% to 66%). The 1:24,000 scale maps were also more time-consuming and costly to produce, due mainly to higher image acquisition costs.
Open the resolved locator · 10.2172/6565521
Metadata licence: OpenAlex metadata (CC0)