-
Notifications
You must be signed in to change notification settings - Fork 0
Image Editing
This page describes the image-level edits that affect display and analysis input.
The current image-edit toolset includes:
- histogram review
- exposure
- contrast
- crop
- uniform exposure
These are image-wide transforms, not per-cell transforms.
The histogram helps you judge the intensity distribution of the current frame.
Use it to answer:
- is the image clipped?
- is the contrast too flat?
- is the region of interest using enough dynamic range?
The histogram should be interpreted together with what you see in the viewer. It is a guide, not the result itself.
Exposure changes the overall brightness of the image.
Use it when:
- the full image is too dark
- the full image is too bright
- you need a clearer review display before annotation or analysis
The current UI blocks mouse-wheel changes on the exposure slider to avoid accidental edits.
Contrast changes the separation between light and dark values.
Use it when:
- event transitions are visually hard to see
- the image has low tonal separation
As with exposure, accidental wheel changes on this control are blocked.
Crop restricts the active image region.
This matters because committed crop affects:
- viewer display
- histogram
- grayscale sampling
- analysis results
If you crop, confirm that every annotated cell remains inside the active area.
Uniform exposure uses a reference area to equalize brightness across frames.
This is useful when:
- illumination drifts over time
- one control region should remain visually consistent
- you need frame-to-frame brightness normalization
Committed image edits affect downstream processing.
That means they are not just cosmetic. They can change the inputs used by measurement and freeze detection.
In practice:
- display changes matter
- histogram changes matter
- analysis changes matter
Image edits are tracked in history.
That means you can:
- preview changes
- commit them
- undo them if needed
- use the smallest edit that makes review and analysis reliable
- avoid over-correcting images
- if you crop, verify the histogram and cell overlay again afterward