Title :
Accuracy and inaccuracy assessments in land-cover classification
Author :
Nishii, Ryuei ; Tanaka, Shojiro
Author_Institution :
Fac. of Integrated Arts & Sci., Hiroshima Univ., Japan
fDate :
1/1/1999 12:00:00 AM
Abstract :
Several measures assessing accuracy of land-cover classification are available, e.g., overall and class-averaged accuracies. Also the kappa statistic is widely used for this purpose. The authors discuss properties of these criteria and point out that the kappa statistic has an unfavorable feature. They propose an alternative coefficient based on Kullback-Leibler information. A test statistic for significance difference between coefficients is also established. Further, the Bayes risk, which takes types of misclassifications into account, is discussed. Their assessment measures are examined through actual error matrices by Z. Ma and R. L. Redmond (1995)
Keywords :
Bayes methods; geophysical signal processing; geophysical techniques; remote sensing; terrain mapping; vegetation mapping; Bayes risk; Bayesian method; Kullback-Leibler information; accuracy; class-averaged accuracies; criteria; error matrices; geophysical measurement technique; inaccuracy assessment; kappa statistic; land surface; land-cover classification; remote sensing; terrain mapping; vegetation mapping; Classification tree analysis; Data analysis; Decision trees; Error analysis; Frequency measurement; Proposals; Statistical analysis; Statistical distributions; Statistics; Testing;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on