Equalize Histogram (OpenCV)

@opencv1 VersionEnhance

Applies CLAHE or global histogram equalization to each Image on a CPU, enhancing the luma channel in a chosen working space (or the whole image in grayscale) and emitting a contrast-enhanced Image.

How it fits

Typical backends

Lift a low-contrast frame before OCR so faint text is more legible.

Brighten a dim camera feed before detection, streaming the result out.

Caveats

I/O contractAn empty input Image passes through unchanged. Color frames are equalized on the luma channel only and keep their channel layout, but when color_space is `gray` (or the input is already single-channel) the output is single-channel, so downstream consumers expecting 3 channels must handle that.
Parameter interactionclip_limit and tile_grid apply only to CLAHE and are ignored when method is `global`; a smaller tile grid produces stronger local contrast and a higher clip limit amplifies contrast but also noise.
AccuracyGlobal equalization is faster than CLAHE but can blow out highlights and shadows; pair with a denoising component upstream if the input is grainy.
State lifetimeAll configuration keys are captured once at startup and the CLAHE operator is built once and reused, so runtime changes have no effect and require a redeploy.

Versions

84680d7bdefaultlatest
Automated release
6/21/2026