At the moment it is quite difficult to compile an OCR workflow from the processors on kwalitee, even if you filter by steps. It would be easier if you could filter the processors according to their position in the OCR workflow (see attached example):
The left side shows all of the OCR steps in a dummy pipeline. When you click on a particular step, the available processors are shown on the right side. The information on the processors is basically as it is now in kwalitee, but in addition there's an example command which you could just copy into ocrd_process, Docker or Taverna.
Processors we recommend for a particular step are highlighted by a (green?) color. This way clickinig through all the steps would provide you with a complete workflow to start with.
At the moment it is quite difficult to compile an OCR workflow from the processors on kwalitee, even if you filter by steps. It would be easier if you could filter the processors according to their position in the OCR workflow (see attached example):
The left side shows all of the OCR steps in a dummy pipeline. When you click on a particular step, the available processors are shown on the right side. The information on the processors is basically as it is now in kwalitee, but in addition there's an example command which you could just copy into ocrd_process, Docker or Taverna.
Processors we recommend for a particular step are highlighted by a (green?) color. This way clickinig through all the steps would provide you with a complete workflow to start with.