Code
How tesseract-ocr/tesseract is laid out, and where to start for common changes.
Imports between files were not traced, so there is no architecture grade.
Where to start reading
- Repository root
- Repository root
- Suggested starting pointAI
- Suggested starting pointAI
- Suggested starting pointAI
- Suggested starting pointAI
- Suggested starting pointAI
- Suggested starting pointAI
Find files for a task
AIAdd a new output format renderer
- Create a new class inheriting from TessBaseRenderer in src/api/renderer.cpp
- Register the new renderer factory in baseapi.cpp's GetRenderers() or similar dispatch method
Add a new character classification feature
- Define feature extraction in src/classify/classify.cpp or create a new module
- Hook feature computation into the LSTM input pipeline in src/lstm/input.cpp
- Retrain language model using src/training/cntraining.cpp with new feature data
Optimize matrix operations for a new CPU target
- Create new dotproduct variant (e.g., src/arch/dotproductavx512.cpp for AVX-512)
- Implement IntSimdMatrix specialization in src/arch/intsimdmatrix.cpp with dispatch logic
- Update CMakeLists.txt in src/training to compile new SIMD variant and select at runtime
Add or update tests
Change build or automation
RepoPilot lists up to 500 files, so a file missing here may still exist.
- No public API or start-up file could be picked out from the file names.