Share Email Print

Proceedings Paper • Open Access

The effects of polynomial interpolation and resampling methods in geometric correction on the land-cover classification accuracy of Landsat-8 OLI imagery: a case study of Kulon Progo area, Yogyakarta

Paper Abstract

Geometric correction is an important step in image pre-processing, because it determines the the positional accuracy of the data. However, the geometric correction also includes pixel values interpolation in their new position, so that it may change original values. This study objectives were (a) to provide information on the effect of geometric correction models on the accuracy of land-cover classification, especially using per-pixel classification with maximum likelihood algorithm; and (b) to assess the effect of image resampling methods on the accuracy of the multispectral classification results. This study made use of Landsat 8 OLI Level 1G imagery covering Kulon Progo Area, Yogyakarta, so that several ground control points (GCPs) were needed to suppress geometric errors. Non-systematic geometric correction was undertaken using first, second and third order polynomial transformations. After that, several resampling processes were applied to the geometrically corrected image, i.e. Nearest Neighbour, Bilinear and Cubic Convolution interpolations. It was found that the affine transformation using six GCPs distributed over the edges of the image, delivered an RMSE value of 0.355539. In addition, the second order polynomial with 10 GCPs scattered around the edges of the image gave an RMSE value of 0.178053. While the third order polynomial transformation with 17 GCPs that were evenly distributed in the image produced an RMSE value of 0.100343. The resampling process produced new images with new pixel values, which were then tested with respect to their classification accuracies based on maximum likelihood algorithm. Samples for accuracy assessment were taken using stratified random sampling strategy. Samples were taken in terms of polygons whose size was determined by considering the pixels’ displacement as the results of geometric corrections. This study also found that resampling with nearest neighbour interpolation using third order polynomial equation produced the best overall accuracy of 75.46%, with a Kappa of 0.7032.

Paper Details

Date Published: 21 November 2019
PDF: 9 pages
Proc. SPIE 11311, Sixth Geoinformation Science Symposium, 113110T (21 November 2019); doi: 10.1117/12.2548423
Show Author Affiliations
Muhammad Zayyanul Afwani, Univ. Gadjah Mada (Indonesia)
Projo Danoedoro, Univ. Gadjah Mada (Indonesia)

Published in SPIE Proceedings Vol. 11311:
Sixth Geoinformation Science Symposium
Sandy Budi Wibowo; Andi B. Rimba; Stuart Phinn; Ammar A. Aziz; Josaphat Tetuko Sri Sumantyo; Hasti Widyasamratri; Sanjiwana Arjasakusuma, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?