Editing the pipeline “recipe”

As mentioned in The data pipeline, data flows through a configurable pipeline. When constructing a complex workflow (e.g. processing different colour channels in different ways - see Isolating a single channel for processing) it can be useful to edit this pipeline directly using the recipe editor (Fig. 15), accessible via the Pipeline Recipe tab. This also lets you access additional functionality such as feature based tracking, which is not currently accessible through the menus. The pipeline is specified by a recipe (a .json textual description) which describes the various processing steps to be applied. The recipe can be modified, either graphically or as text, saved and reloaded on a new dataset, or applied to multiple input files in a batch using the bakeshop utility.

Fig. 15 The pipeline as seen in the recipe editor. Custom workflows can be created by manually adding processing modules.