Implementation of object(s) detection with one or multiple templates in KNIME.
Refer to the wiki section for installation, video tutorial...
The second workflow (nearest template) allows to do classification with a set of templates.
An image thus gets classified as the category of the top-score template, or as outlier if no score is above a user-defined threshold.
The latest workflow version can be downloaded from the KNIME Hub.
In addition to the workflow, a python environment with the Multi-Template-Matching package is necessary.
pip install Multi-Template-Matching (case sensitive)
and finally setup the python envirronment in KNIME.
