Awesome

Vesseract

A "WIP" V wrapper for Tesseract-OCR inspired by pytesseract wrapper!

Tesseract OCR is a optical character recognition software made by Google, it can "read" text from an image.

You need to install Tesseract 4 and register it in your path.

Install the package from VPM:

v install barrack-obama.vesseract

Quickstart

import barrack-obama.vesseract

// Extract text from image
// Image: path - lang: Language for tesseract - args: custom arguments
text := vesseract.image_to_string(
        image: 'sample/demo.png', lang: 'eng', args: '') or {
		panic(err)
}
// Or simply: vesseract.image_to_string_path('sample/demo.png')
// "Got: Hi from Vesseract !"
println("Got: $text")

// Get Tesseract version
version := vesseract.get_tesseract_version() or { panic(err) }
// 4.1.0 installed: "Tesseract 4 - 1 - 0 detected!"
println("Tesseract $version.major - $version.minor - $version.patch detected!")


// Get languages supported by Tesseract
langs := vesseract.get_languages() or { panic(err) }
// Example: "['afr', 'amh', 'ara', 'asm', ... 'uzb_cyrl', 'vie', 'yid', 'yor']"
println("$langs")


// Get alto xml - Require Tesseract >4.1.0
alto := vesseract.image_to_alto_xml_path('sample/demo.png') or { panic(err) }
// "XML: <?xml version="1.0" encoding="UTF-8"?> ... "
println("XML: $alto")


// Get bounding boxes for letters
boxes := image_to_boxes(image: 'sample/demo.png', lang: 'eng', args: '') or { panic(err) }
println("$boxes")
/* "[.Tesseract_box{
    letter: 'H'
    x1: 68
    y1: 206
    x2: 91
    y2: 235
    page: 0
}, ... "


// Check if a language model exists in Tesseract
if (vesseract.is_language_code_supported("fra")) {
    // Do stuff...
} else {
    // Use another model instead...
}

License

This wrapper is licensed under the MIT License, see LICENSE for details