EVALUATION OF TEXTURAL AND MULTIPOLARIZATION RADAR FEATURES FOR CROP CLASSIFICATION

The aim of this research is to evaluate crop discrimination using airb orne radar data based on multipolarization and textural information. M ultipolarization data (C-HH, C-VV, and C-HV) were used for discriminat ing 5 crop types i.e., corn, wheat, soya, pasture, and alfalfa. For th e multipolarization evaluation, an unsupervised classification algorit hm and a supervised method based on maximum likelihood were used on th e data. For the textural evaluation, textural measures of different de grees were calculated on three different order histograms and were eva luated from the crop discrimination point of view. Results show that m ultipolarization correct classification rates of 86.31% and 74.47% wer e obtained for supervised and unsupervised methods respectively. Hence , multipolarization radar data offer an adequate tool for crop identif ication especially with supervised classification. The evaluation of t extural measures shows that for a first order histogram the mean measu re gives the best rate of discrimination. In the case of second and th ird order histograms, the best measures are contrast and large number emphasis respectively. These textural measures were integrated with th e three multipolarization channels in order to determine their specifi c contributions. Results show that crop class separability is thereby improved and that the rate of correct classification increased hy 9.79 % for the crops.