Answers are grounded in this paper and cite their sources.
The main objective of this study was to investigate the role of principal component analysis to improve the accuracy of classification. Methodology: In the present study, PCA has been successfully applied in IRS images of north of Iran (Shafaroud) showing that the first principal components contain more variance of the information in the original four bands. Results: Classification was performed on two sets of data. In the first one, original bands of LISS III (b2-b3-b4-b5) and in the second classification, original bands of LISS III in combination with first component (pca1) was used. Classification was performed with five classes including sea, agriculture, settlements, broad leaf forest and needle leaf forest. In (b2-b3-b4-b5) and b2-b3-b4-b5 in composition with pca1, obtained overall accuracy and kappa coefficient were 99.27-98.94 and 99.37-99.09, respectively. Conclusion: The obtained results indicate that overall accuracy and kappa coefficient increases when pca 1 used along with main bands.
No takes yet. Share an insight, caveat, or question.
Dolati et al. (2016) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: