Change detection of soil states

As I mentioned in a previous post, last September I participated to
the Recent Advances in Quantitative Remotes Sensing Symposium. I
presented several posters there. One of them was about the work done
by Benoît Beguet for his master thesis while he was at CESBIO earlier
this year.

The goal of the work was to assess the potential of high temporal and
spatial resolution multispectral images for the monitoring of soil
states related to agricultural practices.

This is an interesting topic for several reasons, the main ones being:

    1. a bare soil map at any given date is useful for erosion forecast
      and nitrate pollution estimations


  1. the knowledge about the dates of different types of agricultural
    soil work can give clues about the type of crop which is going to
    be grown

The problem was difficult, since we used 8 m. resolution images (so no

useful texture signature is present) and we only had 4 spectral bands
(blue, green, red and near infrared). Without short-wave infra-red
information, it is very difficult to infer something about the early and late
vegetation phases.However, we obtained interesting results for some states and, most of
all, for some transitions – changes – between states.

You can have a look at the poster we presented at RAQRS here