NDVI-based vigour maps production using automatic detection of vine rows in ultra-high resolution aerial images

In this paper, a new method to produce accurate vigour maps using a vine row automatic detection process is presented. It has been applied to 5 cm ground resolution multispectral aerial images (NIR, red, green). NDVI index has been calculated for pre-processed images, obtained by the selection of pixels that represent vine rows, masking non-vine vegetation and other background elements. The adopted pre-processing settlement is constituted by three steps based on dynamic segmentation, Hough Space Clustering and Total Least Squares techniques. The adaptive features of the algorithm make it robust in the presence of inter-row grassing, tree shadows and non-uniform image illumination.