Goko/Tracker
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Chinese OCR 2

Browse Source
This commit is contained in:
MangoPig
2026-02-21 23:04:57 +00:00
parent 7387d17a20
commit 6f2d664185
1 changed files with 4 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Code/TODO.md
+4
View File
@@ -31,6 +31,7 @@ Rationale: De-risk unknowns (Docling OCR quality, Ollama CPU inference speed) be
- [x] Test OCR on images (PNG) — works with `force_ocr=true` - [x] Test OCR on images (PNG) — works with `force_ocr=true`
**Docling API Spec:** **Docling API Spec:**
- Image: `ds4sd/docling-serve:latest` - Image: `ds4sd/docling-serve:latest`
- Port: `5001` - Port: `5001`
- Endpoint: `POST /v1/convert/file` - Endpoint: `POST /v1/convert/file`
@@ -39,12 +40,14 @@ Rationale: De-risk unknowns (Docling OCR quality, Ollama CPU inference speed) be
- OCR engine: `easyocr` (default), supports CJK - OCR engine: `easyocr` (default), supports CJK
**OCR Options:** **OCR Options:**
- `do_ocr=true` — enabled by default, processes bitmap content - `do_ocr=true` — enabled by default, processes bitmap content
- `force_ocr=true` — use for images/scanned PDFs (replaces existing text with OCR) - `force_ocr=true` — use for images/scanned PDFs (replaces existing text with OCR)
- `ocr_engine` — `easyocr` (default), `tesseract`, `rapidocr`, `tesserocr`, `ocrmac` - `ocr_engine` — `easyocr` (default), `tesseract`, `rapidocr`, `tesserocr`, `ocrmac`
- `ocr_lang` — language codes (engine-specific), e.g. `ch_sim` for Simplified Chinese - `ocr_lang` — language codes (engine-specific), e.g. `ch_sim` for Simplified Chinese
**Tested Formats:** **Tested Formats:**
- PDF (Filing Receipt) — extracted text, tables, CJK content ✅ - PDF (Filing Receipt) — extracted text, tables, CJK content ✅
- PNG (Email screenshot) — extracted subject, dates, recipient ✅ (CJK trademark garbled, may need `ocr_lang`) - PNG (Email screenshot) — extracted subject, dates, recipient ✅ (CJK trademark garbled, may need `ocr_lang`)
@@ -56,6 +59,7 @@ Rationale: De-risk unknowns (Docling OCR quality, Ollama CPU inference speed) be
- [x] Validate extraction accuracy (deadline, doc type, etc.) - [x] Validate extraction accuracy (deadline, doc type, etc.)
**Ollama API Spec:** **Ollama API Spec:**
- Image: `ollama/ollama:0.9.3` - Image: `ollama/ollama:0.9.3`
- Port: `11434` - Port: `11434`
- Model: `qwen3:1.7b` (~1.4 GB, auto-pulled on startup) - Model: `qwen3:1.7b` (~1.4 GB, auto-pulled on startup)