PRLib (libprecog)

Pre-Recognition Library.

The main aim of the library: prepare image for text recognition (OCR). Image processing can really help to improve recognition quality.

Features

• Auto crop image
• Background normalization
• Binarization (Global Otsu, Local Otsu, Sauvola, Feng, Niblack, Wolf-Jolion, NICK, FBCITB, Adaptive)
• Blur detection (LAPM, LAPV, TENG, GLVN)
• Color balance
• Crop
• Deblur (Gaussian-based)
• Denoise (Non-local Means Denoising algorithm, Salt-Pepper)
• Deskew
• Find angle and orientation of an image
• Gamma correction
• Hole punch removal
• Line removal
• Perspective warp removal
• Thinning (Zhang-Suen, Guo-Hall)
• White balance (Simple, Grayworld)

Dependencies

• OpenCV
• Leptonica

How to build

1. Go to the directory with source code
2. Make directory for build:

    mkdir build

3. Run cmake:

    cmake ..

4. Run make:

    make

Build tested

• Opencv: 4.7.0
• Leptonica: 1.82.0-3