RETRIEVING FOREST STRUCTURE VARIABLES FROM VERY HIGH RESOLUTION SATELLITE IMAGES USING AN AUTOMATIC METHOD

The main goal of this study is to define a method to describe the forest structure of maritime pine stands from Very High Resolution satellite imagery. The emphasis is placed on the automatisation of the process to identify the most relevant image features, exploiting both spectral and spatial information. Our approach is based on linear regressions between the forest structure variables to be estimated and various spectral and Haralick’s texture features (derived from Grey Level Co-occurrence Matrix). The main drawback of this well- known texture representation is the underlying parameters (window size, displacement length, orientation and quantification level) which are extremely difficult to set due to the spatial complexity of forest structure. To tackle this major issue, probably the main cause of poor texture analysis in practice, we propose an automatic feature selection process whose originality lies on the use of image test frames of adequate forest samples whose forest structure variables were measured at ground. This method, inspired by camera calibration protocols, selects the best image features via statistical modelling, exploring a wide range of parameter values. Hence, just a few samples are required to build up the test frames but allow a fast assessment of thousands of descriptors, given the large number of tested combinations of parameters values. This method was developed and tested on Quickbird panchromatic and multispectral images. It has been successfully applied to the modelling of 7 typical forest structure variables (age, tree height, crown diameter, diameter at breast height, basal area, density and tree spacing). The coefficient of correlation, <i>R</i>², of the best single models for 6 of the forest variables of interest, estimated from the test frames, ranges from 0.89 to 0.97. Only the basal area was weakly correlated to the considered image features (0.64). To improve the results, combinations of panchromatic and or multi-spectral features were tested using multiple linear regressions. As collinearity is a very perturbing problem in multi-linear regression, this issue is carefully addressed. Different variables subset selection methods are tested. A new stepwise method, derived from LARS (Least Angular Regression), turned out the most convincing, significantly improving the quality of estimation for all the forest structure variables (<i>R</i>² > 0:98). Validation is done through stand ages retrieval along the whole site. The best estimation results are obtained from subsets combining multi-spectral and panchromatic features, with various values of window size, highlighting the potential of a multi-scale approach for retrieving forest structure variables from VHR satellite images