Benjamin Admin f9bad7beaa Filter phantom rows from recovered color artifacts and low-conf OCR noise ...

- Apply recovered-artifact filter to ALL zones (was box-zones only)
- Filter any recovered word with text ≤ 2 chars (not just !?•·)
- Add post-grid junk-row removal: rows where all word_boxes have
  conf < 50 and text ≤ 3 chars are dropped as OCR noise

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-18 09:00:43 +01:00

..

backend

Filter phantom rows from recovered color artifacts and low-conf OCR noise

2026-03-18 09:00:43 +01:00

docs

…

frontend

…

scripts

…

Dockerfile

…

Dockerfile.base

feat: OCR pipeline step 8 — validation view with image detection & generation

2026-03-05 10:40:37 +01:00