Vine field monitoring using high resolution remote sensing images: segmentation and characterization of rows of vines

A new framework for the segmentation and characterization of row crops on remote sensing images has been developed and validated for vineyard monitoring. This framework operates on any high-resolution remote sensing images since it is mainly based on geometric information. It aims at obtaining maps describing the variation of a vegetation index such as NDVI along each row of a parcel.
The framework consists in several steps. First, the segmentation step allows the delineation of the parcel under consideration. A region-growing algorithm, based on the textural properties of row crops, was developed for this purpose. Once the parcel under consideration is delineated, a boundary smoothing process is applied and the row detection process begins. Row detection operates by means of an active contour model based on a network of parallel lines. The last step is the design of vegetative vigor maps. Row vigor is computed using pixels neighboring the lines of the network. Once row vigor is obtained on the rows, 2D vigor-maps are constructed. The values measured on the row are propagated to the inter-row pixels, producing «continuous» vigor maps ready to be exported to a GIS software.
We successfully exercised our framework on vineyard images. The resulting parcel segmentations and row detections were accurate and the overall computational time was acceptable.

Authors: Jean-Pierre DA COSTA, Christian GERMAIN, Olivier LAVIALLE, Saeid HOMAYOUNI and Gilbert GRENIER

LAPS CNRS – ENITAB – ENSEIRB, Université Bordeaux 1
351 cours de La Libération, 3305 Talence cedex, France

Email: jean-pierredacosta@laps.u-bordeaux1.fr

Keywords: remote sensing, image processing, row crop, vine

Related Posts

Share via
Copy link
Powered by Social Snap