xtract.bot

CLAHE (adaptive histogram equalize)

Try it interactively →
POST /api/image-clahe

Apply CLAHE — adaptive contrast enhancement that works on local regions instead of globally. Useful for shadowed, low-contrast, or medical images.

CLAHE is a more sophisticated form of histogram equalisation that operates on local tiles instead of the whole image. The "contrast-limited" part prevents amplifying noise in flat regions. Useful for: - Shadowed photos where global tonal correction would blow out highlights. - Medical / scientific imagery with low local contrast. - Pre-processing before OCR or edge detection on uneven documents.

Inputs

NameTypeDefaultDescription
image*fileInput image bytes.
clipLimitnumber (0.5…40)2Contrast clip limit. Higher = more aggressive.
tileSizenumber (2…64)8Side length of each square tile.

Response

Modes: binary, json. Cache: yes (24h TTL).