Kernel-based extreme learning machine for remote-sensing image classification

TLDR

The study evaluates a kernel-based extreme learning machine for land cover classification using multi‑ and hyperspectral remote‑sensing data. The algorithm was benchmarked against support vector machines with a radial basis kernel, assessing parameter simplicity, classification accuracy, and computational cost. The kernel‑based ELM matched or exceeded SVM accuracy, required fewer parameters, and incurred lower computational cost without needing a multiclass strategy. Acknowledgements: data were provided by Prof.

Abstract

Abstract This letter evaluates the effectiveness of a new kernel-based extreme learning machine (ELM) algorithm for a land cover classification using both multi- and hyperspectral remote-sensing data. The results are compared with the most widely used algorithms – support vector machines (SVMs). The results are compared in terms of the ease of use (in terms of the number of user-defined parameters), classification accuracy and computation cost. A radial basis kernel function was used with both the SVM and the kernel-based extreme-learning machine algorithms to ensure compatibility in the comparison of the two algorithms. The results suggest that the new algorithm is similar to, or more accurate than, SVM in terms of classification accuracy, has notable lower computational cost and does not require the implementation of a multiclass strategy. Acknowledgement The ATM data were provided by Prof. Giles Foody of University of Nottingham, UK and were acquired as part of European AgriSAR campaign. The DAIS data were collected and processed by DLR and were kindly made available by Prof. J. Gumuzzio of the Autonomous University of Madrid, Spain. Michael P. Strager, Division of Resource Management, WVU, provided the RapidEye Data and Adam Riley, and natural resource analysis center (NRAC) at WVU helped in processing the LiDAR data. Authors also acknowledge West Virginia department of environmental protection and NRAC for LiDAR data.