A hallmark of most CellProfiler pipelines is the identification of cellular features in your images, whether they are nuclei, organelles or something else.
A number of modules are dedicated to the purpose of detecting these features; the IdentifyPrimaryObjects module is the one that is most commonly used. The result of this module is a set of labeled objects; we define an object as a collection of connected pixels in an image which share the same label. The challenge here is to find a combination of settings that best identify the objects from the image, a task called segmentation. The typical expectation is to end up with one object for each cellular feature of interest (for example, each nucleus is assigned to a single object in a DNA stained image). If this is not the case, the module settings can be adjusted to make it so (or as close as possible). In some cases, image processing modules must be used beforehand to transform the image so it is more amenable to object detection. |
In brief, the workflow of finding objects using this module is to do the following:
Other modules are able to take the results of this module and use them in combination with additional images (like IdentifySecondaryObjects) or other objects (like IdentifyTertiaryObjects) to define yet more objects.
For more information on these identification modules work and how to
configure them for best performance, please see the detailed help by
selecting the IdentifyPrimaryObjects module and clicking the
button at the bottom of the
pipeline panel.