Belongs to the legend of the national base map indeed includes and thus up-to-date data concerning this theme is needed to be able to draw the map. Furthermore, the future national large scale land cover database will also include a "vineyards" layer.
Different databases concerning vineyards produced by several authorities are already available. These external data sources could thus help to produce a "vineyards" layer of this land cover database. Some of these databases are exhaustive ones but only contain the area covered by vineyards per cadastral parcel. On the other hand, other ones offer locally a very fine plotting of vineyards but are not exhaustive at all. As a consequence, a simple integration of these available data sources (and even of the most exhaustives among them) does not seem to be always sufficient to obtain the needed "vineyards" layer directly.
As a consequence, vineyards will sometimes have to be extracted (at least manually and selectively to correct some mistakes) directly from aerial ortho-images to produce this "vineyards" layer of the national land cover database. Such a task should therefore be (at least partially) automated.
Furthermore, it would also be interesting to be able to qualify the available external data sources. A possible way to do this can consists in comparing them to the result of an automatic detection of vineyards from image data, or simply analysing indices used by this detection method. Besides, some vineyards parcels can be missed even in the most exhaustive information sources and would be interesting to be detected (semi-)automatically out of aerial images. As a consequence, there is a need for semi-automatic detection of vineyards out of aerial images.
Vineyards can have different aspects on aerial images. This phenomenon depends on various parameters (season, age of the plants, different ways of culture...). Nevertheless, vineyards always present a high frequency periodic texture. However, it must here be said that orchards also have a periodic texture but at a slightly lower frequency.
In this paper, two distinct approaches are proposed and tested to achieve this (semi-)automatic detection of vineyards task out of 50cm ground resolution ortho-images from the national ortho-image database. Both are object based approaches relying on image texture analysis in homogeneous land cover regions.
Therefore, the first step (common to both approaches) is a segmentation of the image into homogeneous land cover regions. These regions can then be classified are vineyards or not by the following approaches.
A first approach consists in a frequential analysis of the image texture in each region. A semi-variogram is first calculated from ortho-image for each region of the segmentation. A Fourier transform (FFT) of this semi-variogram of the image is then considered. If a periodic signal with a high frequency (i.e. of which the frequency is upper than a threshold) is identified, the region is labeled as a vineyard.
The second approach is a supervised (per region) land cover classification one. It uses texture indices calculated from ortho-images as input image information. In particular, some texture indices derived from SIFT descriptors calculated from ortho-images have been used in the experiments, giving good results.
Results obtained by these two approaches can then be merged.