Development and application of a machine learning-based method for the automated identification of soil horizons from photos of soil profiles.

Development and application of a machine learning-based method for the automated identification of soil horizons and some of their morphological properties (such as rust spotting, wash-in features) from photos of soil profiles. Image processing and pattern recognition techniques will be used to segment and classify the ground horizons and provide valid results. In addition to the overall analysis of the profile, the respective individual horizons are also to be identified.