Commit Graph

17 Commits

Author SHA1 Message Date
Benjamin Admin
164b35c06a fix(ocr-pipeline): tighten page_ref constraints based on live testing
- Reduce left-side threshold from 35% to 20% of content width
- Strong language signal (eng/deu > 0.3) now prevents page_ref assignment
- Increase column_ignore word threshold from 3 to 8 for edge columns
- Apply language guard to Level 1 and Level 2 classification

Fixes: column with deu=0.921 was misclassified as page_ref because
reference score check ran before language analysis.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 23:33:11 +01:00
Benjamin Admin
db8327f039 fix(ocr-pipeline): tune column detection based on GT comparison
Address 5 weaknesses found via ground-truth comparison on session df3548d1:
- Add column_ignore for edge columns with < 3 words (margin detection)
- Absorb tiny clusters (< 5% width) into neighbors post-merge
- Restrict page_ref to left 35% of content area across all 3 levels
- Loosen marker thresholds (width < 6%, words <= 15) and add strong
  marker score for very narrow non-edge columns (< 4%)
- Add EN/DE position tiebreaker when language signals are both weak

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 23:16:31 +01:00
Benjamin Admin
587b066a40 feat(ocr-pipeline): ground-truth comparison tool for column detection
Side-by-side view: auto result (readonly) vs GT editor where teacher
draws correct columns. Diff table shows Auto vs GT with IoU matching.
GT data persisted per session for algorithm tuning.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 22:48:37 +01:00
Benjamin Admin
03fa186fec fix(ocr-pipeline): increase merge distance to 6% for better column merging
Sub-alignments within a column (indented words, etc.) were 60-90px apart
and not getting merged at 3%. On a typical 5-col page (~1500px), 6% = ~90px
merges sub-alignments while keeping real column boundaries (~300px) separate.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 20:19:09 +01:00
Benjamin Admin
1040729874 fix(ocr-pipeline): avoid backslash in f-string for Python 3.11 compat
Use format() instead of nested f-strings with escaped quotes.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 20:06:20 +01:00
Benjamin Admin
4f37afa222 feat(ocr-pipeline): verticality filter for column detection
Clusters now track Y-positions of their words and filter by vertical
coverage (>=30% primary, >=15%+5words secondary) to reject noise from
indentations or page numbers. Merge distance widened to 3% content width.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 19:57:13 +01:00
Benjamin Admin
bb879a03a8 feat(ocr-pipeline): add column_ignore type for margins/empty areas
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-27 08:51:56 +01:00
Benjamin Admin
1393a994f9 Flexible inhaltsbasierte Spaltenerkennung (2-Phasen)
Ersetzt hardcodierte Positionsregeln durch ein zweistufiges System:
Phase A erkennt Spaltengeometrie (Clustering), Phase B klassifiziert
Typen per Inhalt (Sprache/Rolle) mit 3-stufiger Fallback-Kette.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 23:33:35 +01:00
Benjamin Admin
cf27a95308 feat(ocr-pipeline): word-based 5-column detection for vocabulary pages
Replace projection-profile layout analysis with Tesseract word bounding
box clustering to detect 5-column vocabulary layouts (page_ref, EN, DE,
markers, examples). Falls back to projection profiles when < 3 clusters.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 23:08:14 +01:00
Benjamin Admin
aa06ae0f61 feat: Persistente Sessions (PostgreSQL) + Spaltenerkennung (Step 3)
Sessions werden jetzt in PostgreSQL gespeichert statt in-memory.
Neue Session-Liste mit Name, Datum, Schritt. Sessions ueberleben
Browser-Refresh und Container-Neustart. Step 3 nutzt analyze_layout()
fuer automatische Spaltenerkennung mit farbigem Overlay.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 22:16:37 +01:00
Benjamin Admin
09b820efbe refactor(dewarp): replace displacement map with affine shear correction
The old displacement-map approach shifted entire rows by a parabolic
profile, creating a circle/barrel distortion. The actual problem is
a linear vertical shear: after deskew aligns horizontal lines, the
vertical column edges are still tilted by ~0.5°.

New approach:
- Detect shear angle from strongest vertical edge slope (not curvature)
- Apply cv2.warpAffine shear to straighten vertical features
- Manual slider: -2.0° to +2.0° in 0.05° steps
- Slider initializes to auto-detected shear angle
- Ground truth question: "Spalten vertikal ausgerichtet?"

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 18:23:04 +01:00
Benjamin Admin
ff2bb79a91 fix(dewarp): change manual slider to percentage (0-200%) instead of raw multiplier
The old -3.0 to +3.0 scale multiplied the full displacement map (up to ~79px)
directly, causing extreme distortion at values >1. New slider:
- 0% = no correction
- 100% = auto-detected correction (default)
- 200% = double correction
- Step size: 5%

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 18:10:34 +01:00
Benjamin Admin
fb496c5e34 perf(klausur-service): split Dockerfile into base + app layer
Tesseract OCR + 70 Debian packages + pip dependencies are now in a
separate base image (klausur-base:latest) that is built once and reused.
A --no-cache build now only rebuilds the code layer (~seconds) instead
of re-downloading 33 MB of system packages (~9 minutes).

Rebuild base when requirements.txt or system deps change:
  docker build -f klausur-service/Dockerfile.base -t klausur-base:latest klausur-service/

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 17:43:24 +01:00
Benjamin Admin
9df745574b fix(ocr-pipeline): dewarp visibility, grid on both sides, session persistence
- Fix dewarp method selection: prefer methods with >5px curvature over
  higher confidence (vertical_edge 79px was being ignored for text_baseline 2px)
- Add grid overlay on left image in Dewarp step for side-by-side comparison
- Add GET /sessions/{id} endpoint to reload session data
- StepDeskew accepts sessionId prop to restore state when navigating back
- SessionInfo type extended with optional deskew_result and dewarp_result

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 17:29:53 +01:00
Benjamin Admin
589d2f811a feat: Dewarp-Korrektur als Schritt 2 in OCR Pipeline (7 Schritte)
Implementiert Buchwoelbungs-Entzerrung mit zwei Methoden:
- Methode A: Vertikale-Kanten-Analyse (Sobel + Polynom 2. Grades)
- Methode B: Textzeilen-Baseline (Tesseract + Baseline-Kruemmung)
Beste Methode wird automatisch gewaehlt, manueller Slider (-3 bis +3).

Backend: 3 neue Endpoints (auto/manual dewarp, ground truth)
Frontend: StepDewarp + DewarpControls, Pipeline von 6 auf 7 Schritte

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 16:46:41 +01:00
Benjamin Admin
d552fd8b6b feat: OCR Pipeline mit 6-Schritt-Wizard fuer Seitenrekonstruktion
All checks were successful
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-school (push) Successful in 38s
CI / test-go-edu-search (push) Successful in 29s
CI / test-python-klausur (push) Successful in 1m46s
CI / test-python-agent-core (push) Successful in 17s
CI / test-nodejs-website (push) Successful in 22s
Neue Route /ai/ocr-pipeline mit schrittweiser Begradigung (Deskew),
Raster-Overlay und Ground Truth. Schritte 2-6 als Platzhalter.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 15:38:08 +01:00
Benjamin Boenisch
5a31f52310 Initial commit: breakpilot-lehrer - Lehrer KI Platform
Services: Admin-Lehrer, Backend-Lehrer, Studio v2, Website,
Klausur-Service, School-Service, Voice-Service, Geo-Service,
BreakPilot Drive, Agent-Core

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 23:47:26 +01:00