Spotlights

The spotlights will focus on topics related to the compilation of a comprehensive image dataset suitable to use as a basis for further environmental data analysis.

Control files and functions

At the moment there don’t seem to be any real advantages to using functions instead of control scripts. The main reason is that we just created a batch file that simply processes external functions (e.g. those of the lidR package) one after the other.

Spotlight: Artificial Images

Artificial images result from mathematical operations on image data. The operations are pixel or neighborhood based.

Spotlight: Land-cover Classification

Classifications aim in grouping pixel values into homogeneous categories. The categories can be defined a priori or posteriori whichever supervised or unsupervised classification model is used.

Spotlight: Variable selection and prediction

The appropriate collection of training data is essential for all monitored classifications. In theory, the rules to be followed are simple, but it is often difficult to achieve a robust, practical application.