Relationships of image classification accuracy and variation of landscape statistics
Shao, G; Liu, D; Zhao, G
AbstractMultiple classifications of a Landsat Thematic Mapper (TM) image resulted in 23 thematic maps for an area of 31,660 ha in central Indiana in the United States. Each map consists of four land use and land cover types (classes): urban, agricultural land, forest, and water A common set of reference data was randomly sampled from the image and was used to evaluate the classification accuracy of each of the 23 maps. The classification accuracy of these maps was between 77.6 to 89.2%. Landscape index values for nine landscape-level indices and nine class-level indices were derived from the 23 maps. Although the majority of the thematic maps were not significantly, different in overall accuracy, the landscape index values showed high variation among the maps. The results suggested that the variation of landscape index values is inversely, proportional to classification accuracy. Therefore, when classification accuracy is lower, the uncertainties of landscape characterization become higher Unfortunately, such uncertainties are not readily, predictable because landscape index values can be altered by the magnitude and spatial distribution of classification errors. Classification accuracy alone is not sufficient to ensure the accuracy of landscape characterization.
WOS Research AreaRemote Sensing
WOS SubjectRemote Sensing
WOS IDWOS:000172767800005
Citation statistics
Document Type期刊论文
Corresponding AuthorShao, G
Affiliation1.Purdue Univ, Dept Forestry & Nat Resources, W Lafayette, IN 47907 USA
2.Chinese Acad Sci, Inst Appl Math, Beijing 100080, Peoples R China
3.S Carolina Dept Hlth & Environm Control, Div Biostat, Columbia, SC 29201 USA
Recommended Citation
GB/T 7714
Shao, G,Liu, D,Zhao, G. Relationships of image classification accuracy and variation of landscape statistics[J]. CANADIAN JOURNAL OF REMOTE SENSING,2001,27(1):35-45.
APA Shao, G,Liu, D,&Zhao, G.(2001).Relationships of image classification accuracy and variation of landscape statistics.CANADIAN JOURNAL OF REMOTE SENSING,27(1),35-45.
MLA Shao, G,et al."Relationships of image classification accuracy and variation of landscape statistics".CANADIAN JOURNAL OF REMOTE SENSING 27.1(2001):35-45.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Shao, G]'s Articles
[Liu, D]'s Articles
[Zhao, G]'s Articles
Baidu academic
Similar articles in Baidu academic
[Shao, G]'s Articles
[Liu, D]'s Articles
[Zhao, G]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Shao, G]'s Articles
[Liu, D]'s Articles
[Zhao, G]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.