CarrotOCR

Implementation of OCR from open sources for internal tasks of the Tatneft Group.

Project Organization

├── LICENSE
├── README.md               <- The top-level README for developers using this project.
│
├── docs                    <- A default Sphinx project; see sphinx-doc.org for details
│
├── notebooks               <- Jupyter notebooks. Example of the model usage.
│
├── requirements.txt 
│
├── setup.py                <- makes project pip installable (pip install -e .) so src can be imported
├── src                     <- Source code for use in this project.
│   ├── api                 
│   │   ├── crud            
│   │   ├── models          <- Fast API models.
│   │   ├── endpoints       <- Fast API endpoints.
│   │   └── main.py
│   │
│   ├── db                  <- Scripts that using for connect to the database, 'processed' and 'permatags' table manager.
│   │   ├── database_manager.py
│   │   ├── processed_structure.py
│   │   ├── processed_manager.py
│   │   ├── permatags_manager.py
│   │   └── permatags_manager.py
│   │
│   │
│   ├── features            <- Scripts to turn raw data into features for modeling
│   │   ├── build_features.py
│   │   └── extract_features.py
│   │
│   ├── ai_models           <- Scripts with loading trained models.
│   │   ├── ocr_models      <- Factory Method pattern.
│   │   |   ├── none_ocr.py       
│   │   |   ├── ocr.py       
│   │   |   ├── easyocr.py  
│   │   |   └── pytesseract.py 
│   │   |   
│   │   ├── weights         <- Weights of models
│   │   └── find_tags.py
│   │
│   ├── tests               <- Unittest for each class in src.
│   │
│   ├── visualization       <- Scripts to create exploratory and results oriented visualizations
│   │   └── visualize.py
│   │
│   └── utils               <- Scripts that contains some utils.
│       └── utils.py
│
│
└── tox.ini                 <- tox file with settings for running tox; see tox.readthedocs.io

---

Run project

Building Docker Image

It is preferable to run the project on the linux

 docker build -t carrot-ocr . --no-cache  2>&1 | tee build_log.txt

FastAPI Swagger

Now you can open locally this site http://127.0.0.1:8000/api/docs.

Example of usage

import cv2
from src.visualization.visualize import *
from src.ai_models.ocr import OCRModelFactory

url = 'data/raw/example1.jpg'
image = cv2.imread(url)

model = OCRModelFactory.get("easyocr")
outputs = model([image])
result2show(image, outputs[0])