# ROI: Measurements

ROI objects define regions of interest for image measurements. The ROI is a
core element in analyzer analysis and is required for analyzer construction.
For analysis that does not require an ROI, a placeholder `NoROI` is used. In
particular, `LineROI` is used to extract line profiles from images, and area
ROI classes are used to extract area information or perform other types of
analysis.

The ROI class can be constructed using an ImageJ ROI file or manually specified
parameters.

## Example

```python
import numpy as np

from pyleem.roi import LineROI, RectROI

image = np.arange(10000).reshape(100, 100)

# Line ROI: extract a line profile and intensity statistics.
line_roi = LineROI(src=(10, 5), dst=(10, 80), linewidth=3)
line_measurement = line_roi.measure(image)
profile = line_measurement.profile

# Area ROI: measure intensity statistics inside a rectangle.
area_roi = RectROI(top=20, left=30, bottom=50, right=60)
area_measurement = area_roi.measure(image)
mean_intensity = area_measurement.mean

# Save and reload ImageJ ROI files.
line_roi.tofile("line.roi")
area_roi.tofile("area.roi")
line_roi = LineROI(roi_file="line.roi")
```


- {doc}`../api/roi`: For custom ROI implementations.
<!-- - {doc}`../ref/roi_ref`: For full module details. -->