# tesseract-ocr/tesseract architecture

> Architecture diagram of the GitHub repository tesseract-ocr/tesseract, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-11-02). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/tesseract-ocr/tesseract
- Repository: https://github.com/tesseract-ocr/tesseract

## Overview

Tesseract is an OCR engine exposed as both a command-line program and a C/C++ library. The main user workflow is to provide an image and traineddata, run recognition, and obtain text or a structured output format. Its domain stages include image preparation, layout analysis, recognition, language-aware selection, and rendering; training is a separate major capability. Relationships are limited to README statements and SOURCE INDEX references, so some implementation-level transitions remain unverified.

Main entry points: [`src/api/baseapi.cpp`](https://github.com/tesseract-ocr/tesseract/blob/main/src/api/baseapi.cpp) (C++ OCR API), [`src/ccstruct/pageres.cpp`](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/pageres.cpp) (Page results), [`src/tesseract.cpp`](https://github.com/tesseract-ocr/tesseract/blob/main/src/tesseract.cpp) (CLI runner), [`src/ccstruct/image.h`](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/image.h) (Image input)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_interfaces["User interfaces"]
  node_cli["CLI runner<br/>[tesseract.cpp]"]
  node_library["C++ OCR API<br/>[baseapi.cpp]"]
  node_capi["C OCR API<br/>[capi.cpp]"]
end

subgraph group_recognition["OCR processing"]
  node_image["Image input<br/>[image.h]"]
  node_threshold["Image thresholding<br/>[thresholder.h]"]
  node_layout["Page layout<br/>[colpartition.cpp]"]
  node_segmentation["Page segmentation<br/>[pagesegmain.cpp]"]
  node_page_state["Page results<br/>[pageres.cpp]"]
  node_lstm["LSTM recognition<br/>[lstmrecognizer.cpp]"]
  node_legacy["Legacy classifier<br/>[classify.cpp]"]
  node_wordrec["Word recognition<br/>[segsearch.cpp]"]
  node_paragraphs["Paragraph detection<br/>[paragraphs.cpp]"]
end

subgraph group_language["Language and data"]
  node_language_model["Language model<br/>[language_model.cpp]"]
  node_dictionary["Dictionary checks<br/>[dict.cpp]"]
  node_unicode["Unicode character set<br/>[unicharset.h]"]
  node_traineddata["Traineddata loader"]
end

subgraph group_outputs["Results and training"]
  node_renderer["Output renderers<br/>[pagerenderer.cpp]"]
  node_result_iterators["Result iterators<br/>[resultiterator.cpp]"]
  node_training["Training tools<br/>[text2image.cpp]"]
end

node_user(("Image user"))
node_leptonica["Leptonica"]

node_user -->|"runs OCR"| node_cli
node_user -->|"calls API"| node_library
node_cli -.->|"configures engine"| node_library
node_cli -.->|"loads image"| node_image
node_image -->|"opens formats"| node_leptonica
node_library -->|"uses results"| node_page_state
node_library -->|"uses language data"| node_dictionary
node_library -->|"uses detection"| node_paragraphs
node_library -.->|"loads data"| node_traineddata
node_capi -.->|"exposes API"| node_library
node_threshold -.->|"prepares layout"| node_layout
node_layout -.->|"analyzes page"| node_segmentation
node_segmentation -.->|"records regions"| node_page_state
node_page_state -.->|"recognizes lines"| node_lstm
node_page_state -.->|"recognizes patterns"| node_legacy
node_wordrec -->|"updates state"| node_language_model
node_language_model -.->|"checks words"| node_dictionary
node_unicode -.->|"defines characters"| node_traineddata
node_page_state -.->|"provides page rows"| node_paragraphs
node_library -.->|"renders output"| node_renderer
node_renderer -->|"writes formats"| node_user
node_library -.->|"returns results"| node_result_iterators
node_user -->|"trains languages"| node_training

click node_cli "https://github.com/tesseract-ocr/tesseract/blob/main/src/tesseract.cpp"
click node_library "https://github.com/tesseract-ocr/tesseract/blob/main/src/api/baseapi.cpp"
click node_capi "https://github.com/tesseract-ocr/tesseract/blob/main/src/api/capi.cpp"
click node_image "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/image.h"
click node_threshold "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/thresholder.h"
click node_layout "https://github.com/tesseract-ocr/tesseract/blob/main/src/textord/colpartition.cpp"
click node_segmentation "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/pagesegmain.cpp"
click node_page_state "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/pageres.cpp"
click node_lstm "https://github.com/tesseract-ocr/tesseract/blob/main/src/lstm/lstmrecognizer.cpp"
click node_legacy "https://github.com/tesseract-ocr/tesseract/blob/main/src/classify/classify.cpp"
click node_wordrec "https://github.com/tesseract-ocr/tesseract/blob/main/src/wordrec/segsearch.cpp"
click node_paragraphs "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/paragraphs.cpp"
click node_language_model "https://github.com/tesseract-ocr/tesseract/blob/main/src/wordrec/language_model.cpp"
click node_dictionary "https://github.com/tesseract-ocr/tesseract/blob/main/src/dict/dict.cpp"
click node_unicode "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccutil/unicharset.h"
click node_traineddata "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccutil/tessdatamanager.cpp"
click node_renderer "https://github.com/tesseract-ocr/tesseract/blob/main/src/api/pagerenderer.cpp"
click node_result_iterators "https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/resultiterator.cpp"
click node_training "https://github.com/tesseract-ocr/tesseract/blob/main/src/training/text2image.cpp"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_cli,node_library,node_capi,node_user toneBlue
class node_image,node_threshold,node_layout,node_segmentation,node_page_state,node_lstm,node_legacy,node_wordrec,node_paragraphs toneAmber
class node_language_model,node_dictionary,node_unicode,node_traineddata toneMint
class node_renderer,node_result_iterators,node_training toneRose
class node_leptonica toneIndigo
```

## Components

### User interfaces

- [CLI runner](https://github.com/tesseract-ocr/tesseract/blob/main/src/tesseract.cpp): `src/tesseract.cpp`
- [C++ OCR API](https://github.com/tesseract-ocr/tesseract/blob/main/src/api/baseapi.cpp): `src/api/baseapi.cpp`
- [C OCR API](https://github.com/tesseract-ocr/tesseract/blob/main/src/api/capi.cpp): `src/api/capi.cpp`

### OCR processing

- [Image input](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/image.h): `src/ccstruct/image.h`
- [Image thresholding](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/thresholder.h): `src/ccmain/thresholder.h`
- [Page layout](https://github.com/tesseract-ocr/tesseract/blob/main/src/textord/colpartition.cpp): `src/textord/colpartition.cpp`
- [Page segmentation](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/pagesegmain.cpp): `src/ccmain/pagesegmain.cpp`
- [Page results](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccstruct/pageres.cpp): `src/ccstruct/pageres.cpp`
- [LSTM recognition](https://github.com/tesseract-ocr/tesseract/blob/main/src/lstm/lstmrecognizer.cpp): `src/lstm/lstmrecognizer.cpp`
- [Legacy classifier](https://github.com/tesseract-ocr/tesseract/blob/main/src/classify/classify.cpp): `src/classify/classify.cpp`
- [Word recognition](https://github.com/tesseract-ocr/tesseract/blob/main/src/wordrec/segsearch.cpp): `src/wordrec/segsearch.cpp`
- [Paragraph detection](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/paragraphs.cpp): `src/ccmain/paragraphs.cpp`

### Language and data

- [Language model](https://github.com/tesseract-ocr/tesseract/blob/main/src/wordrec/language_model.cpp): `src/wordrec/language_model.cpp`
- [Dictionary checks](https://github.com/tesseract-ocr/tesseract/blob/main/src/dict/dict.cpp): `src/dict/dict.cpp`
- [Unicode character set](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccutil/unicharset.h): `src/ccutil/unicharset.h`
- [Traineddata loader](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccutil/tessdatamanager.cpp): `src/ccutil/tessdatamanager.cpp`

### Results and training

- [Output renderers](https://github.com/tesseract-ocr/tesseract/blob/main/src/api/pagerenderer.cpp): `src/api/pagerenderer.cpp`
- [Result iterators](https://github.com/tesseract-ocr/tesseract/blob/main/src/ccmain/resultiterator.cpp): `src/ccmain/resultiterator.cpp`
- [Training tools](https://github.com/tesseract-ocr/tesseract/blob/main/src/training/text2image.cpp): `src/training/text2image.cpp`

### Other components

- **Image user**
- **Leptonica**

## Connections

- Image user → CLI runner: runs OCR (evidence: `README.md`)
- Image user → C++ OCR API: calls API (evidence: `README.md`)
- CLI runner → C++ OCR API: configures engine
- CLI runner → Image input: loads image
- Image input → Leptonica: opens formats (evidence: `README.md`)
- C++ OCR API → Page results: uses results (evidence: `src/api/baseapi.cpp`)
- C++ OCR API → Dictionary checks: uses language data (evidence: `src/api/baseapi.cpp`)
- C++ OCR API → Paragraph detection: uses detection (evidence: `src/api/baseapi.cpp`)
- C++ OCR API → Traineddata loader: loads data
- C OCR API → C++ OCR API: exposes API
- Image thresholding → Page layout: prepares layout
- Page layout → Page segmentation: analyzes page
- Page segmentation → Page results: records regions
- Page results → LSTM recognition: recognizes lines
- Page results → Legacy classifier: recognizes patterns
- Word recognition → Language model: updates state (evidence: `src/wordrec/segsearch.cpp`)
- Language model → Dictionary checks: checks words
- Unicode character set → Traineddata loader: defines characters
- Page results → Paragraph detection: provides page rows
- C++ OCR API → Output renderers: renders output
- Output renderers → Image user: writes formats (evidence: `README.md`)
- C++ OCR API → Result iterators: returns results
- Image user → Training tools: trains languages (evidence: `README.md`)
