Objects within Objects
(General)
Analyzing objects fully enclosed within larger parent objects is a common task in materials and life sciences, where key metrics like count or mean intensity per parent are often critical.
The Image-Pro Objects within Objects protocol provides a flexible, stepwise process for segmenting parent compartments and enclosed objects using thresholding, machine learning, or pre-trained deep learning models. This approach supports unlimited object classes and enables efficient analysis of large, complex datasets and image folders, even for users with little to no image analysis experience.
Techniques: Brightfield, Fluorescence, Holotomography, SEM, TEM
How it works
Select Channels
Select the channel for the large parent compartment and the channels for the smaller child objects.
Find Parent Objects
Find parent objects with a pre-trained deep learning model, machine learning, or threshold segmentation.
Find Child Objects
Find child objects using a pre-trained deep learning model, machine learning, or threshold segmentation.
Quantitative results
Automatically generate tables, heat maps, charts and even complex bespoke reports.
Measurement parameters supported
- • Number of Parent Compartments
- • Number of Objects contained within Parents (counted per class)
- • Mean Parent Intensity
- • Mean Intensity of Child Classes
- • Mean Percentage Parent Area of Each Child Class
- • Custom user-defined measurements
Solution requirements
Required Modules
