ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume II-5/W2
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5/W2, 259–264, 2013
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5/W2, 259–264, 2013

  16 Oct 2013

16 Oct 2013

Non-linear methods for inferring lidar metrics using SPOT-5 textural data

A. Shamsoddini1,2, J. C. Trinder1, and R. Turner3 A. Shamsoddini et al.
  • 1School of Civil and Environmental Engineering, University of New South Wales, Sydney, Australia
  • 2Department of Remote Sensing & GIS, Tarbiat Modares University, Iran
  • 3Remote Census, Sydney, Australia

Keywords: SPOT, Lidar, Forestry, Texture, Neural networks

Abstract. Although many studies have demonstrated the utility of airborne lidar for forest inventory, the acquisition and processing of the data can be cost prohibitive for small areas. In such cases, it may be possible to emulate lidar metrics using more affordable optical data. This study explored processing methods for predicting lidar metrics using SPOT-5 textural data. Multiple-linear regression (MLR) was compared with non-linear machine learning techniques including multi-layer perceptron (MLP) artificial neural networks (ANN), rational basis function (RBF) ANN and regression tree (RT). For this purpose, 11 grey level co-occurrence matrix (GLCM) indices were calculated for bands, band ratios and principal components (PCs) of SPOT-5 multispectral image. SPOT-5 metrics were correlated with 25 lidar metrics collected over a Pinus radiata plantation. After dimensionality reduction, random forest feature selection was applied to select the most relevant SPOT-5 textural attributes for inferring each lidar metric. The results showed that the non-linear methods including MLP and RBF methods are more promising for modelling lidar metrics using SPOT-5 data than MLR and RT.