Accuracy assessment of digitized and classified land cover data for wildlife habitat

GlobeLand30 is the world’s first 30m high resolution land cover data set (Chen et al. 2014) and has been a successful model of Big-Data mining from a host of Landsat imagery, thereby contributing to and enhancing the existing global geospatial knowledge base (GlobeLand30 2014). As there is a lot of uncertainty and errors in the global land cover data, therefore it becomes very difficult to validate land cover on a global scale. Efforts on validating Globeland30 data have been made in various parts of the world in the past and will continue to be done. The objective of this project is to validate GlobeLand30 data set by carrying out a case study in Ontario, Canada. The adopted methodology for doing validation is by using cell-to-cell benchmarking (Maria et al. 2015), thereby deriving Error Matrix, and its derivatives, which includes overall accuracy, user accuracy, producer accuracy and kappa coefficient. The results show that an overall accuracy of 84.14% is obtained for GlobeLand30 data with consideration of shadows, which is relatively a high percentage number indicating that the GlobeLand30 data classification is highly accurate for Ontario, Canada. Keywords: land cover; GlobeLand30; accuracy assessment; Ontario

Download Full-text

Accuracy assessment of Globeland30: a case study of Ontario,Canada

10.32920/ryerson.14647026.v1 ◽

2021 ◽

Author(s):

Rishita Rangarh

Keyword(s):

Land Cover ◽

Accuracy Assessment ◽

Landsat Imagery ◽

Global Scale ◽

Data Set ◽

Error Matrix ◽

Successful Model ◽

Land Cover Data ◽

Made In

GlobeLand30 is the world’s first 30m high resolution land cover data set (Chen et al. 2014) and has been a successful model of Big-Data mining from a host of Landsat imagery, thereby contributing to and enhancing the existing global geospatial knowledge base (GlobeLand30 2014). As there is a lot of uncertainty and errors in the global land cover data, therefore it becomes very difficult to validate land cover on a global scale. Efforts on validating Globeland30 data have been made in various parts of the world in the past and will continue to be done. The objective of this project is to validate GlobeLand30 data set by carrying out a case study in Ontario, Canada. The adopted methodology for doing validation is by using cell-to-cell benchmarking (Maria et al. 2015), thereby deriving Error Matrix, and its derivatives, which includes overall accuracy, user accuracy, producer accuracy and kappa coefficient. The results show that an overall accuracy of 84.14% is obtained for GlobeLand30 data with consideration of shadows, which is relatively a high percentage number indicating that the GlobeLand30 data classification is highly accurate for Ontario, Canada. Keywords: land cover; GlobeLand30; accuracy assessment; Ontario

Download Full-text

The Attribute Accuracy Assessment of Land Cover Data in the National Geographic Conditions Survey

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprsannals-ii-4-35-2014 ◽

2014 ◽

Vol II-4 ◽

pp. 35-40 ◽

Cited By ~ 5

Author(s):

X. Ji ◽

X. Niu

Keyword(s):

Land Cover ◽

Accuracy Assessment ◽

Classification Result ◽

Error Matrix ◽

National Geographic ◽

Land Cover Data ◽

Object Based ◽

Data Production ◽

Traditional Approaches

With the widespread national survey of geographic conditions, object-based data has already became the most common data organization pattern in the area of land cover research. Assessing the accuracy of object-based land cover data is related to lots of processes of data production, such like the efficiency of inside production and the quality of final land cover data. Therefore，there are a great deal of requirements of accuracy assessment of object-based classification map. Traditional approaches for accuracy assessment in surveying and mapping are not aimed at land cover data. It is necessary to employ the accuracy assessment in imagery classification. However traditional pixel-based accuracy assessing methods are inadequate for the requirements. The measures we improved are based on error matrix and using objects as sample units, because the pixel sample units are not suitable for assessing the accuracy of object-based classification result. Compared to pixel samples, we realize that the uniformity of object samples has changed. In order to make the indexes generating from error matrix reliable, we using the areas of object samples as the weight to establish the error matrix of object-based image classification map. We compare the result of two error matrixes setting up by the number of object samples and the sum of area of object samples. The error matrix using the sum of area of object sample is proved to be an intuitive, useful technique for reflecting the actual accuracy of object-based imagery classification result.

Download Full-text