Crate tesseract_ocr_static

Expand description

This crate provides ergonomic Rust interface for underlying Tesseract OCR library. There are two main structs: TextRecognizer and LayoutAnalyzer. TextRecognizer allows one to recognize text from the picture and outputs the text, the bounding boxes and other parameters. LayoutAnalyzer allows one to analyze the layout without recognizing text; it should consume less memory than TextRecognizer.

Pictures can be loaded into the library via Image struct that accepts images in raw RGB/RGBA formats; no other formats are supported. If you need to read an image from a file in another format, you can do so with any Rust crate (e.g. image).

§Examples

§Simple character recognition

use tesseract_ocr_static::{Image, TextRecognizer};
use image::ImageReader;

let rgb = ImageReader::open("hello.txt").unwrap().decode().unwrap().into_rgb8();
let image = Image::from_rgb(rgb.width(), rgb.height(), rgb.as_raw()).unwrap();
let mut recognizer = TextRecognizer::new().unwrap();
let results = recognizer.recognize_text(&image).unwrap();
assert_eq!("Hello world", results.get_utf8_text().as_str());

§Print recognized text and corresponding bounding boxes

use tesseract_ocr_static::{Image, LayoutLevel, TextRecognizer};
use image::ImageReader;

let rgb = ImageReader::open("hello.txt").unwrap().decode().unwrap().into_rgb8();
let image = Image::from_rgb(rgb.width(), rgb.height(), rgb.as_raw()).unwrap();
let mut recognizer = TextRecognizer::new().unwrap();
let results = recognizer.recognize_text(&image).unwrap();
let mut iter = results.iter();
while let Some(word) = iter.next(LayoutLevel::Word) {
    println!(
        "Word {:?}, confidence {:.1}, bounding box {:?}",
        word.get_utf8_text(LayoutLevel::Word).as_str(),
        word.confidence(LayoutLevel::Word),
        word.bounding_box(LayoutLevel::Word),
    );
}

§Build customization

Please consult tesseract-ocr-static-c crate’s documentation.

Structs§

ChoiceIterator: An iterator over classifier choices for a symbol.
ClassifierChoice: A symbol choice.
Config: OCR configuration.
Element: Text layout element.
FontAttrs: Font attributes.
Image: Tesseract-specific image.
InitFailed: Initialization failed
InvalidImage: Invalid image
InvalidVariable: Invalid variable
LayoutAnalyzer: Layout analysis engine.
LayoutIter: An iterator over text layout elements.
Line: Line.
Monitor: Monitor tracks text recognition progress and can be used to set the timeout.
OrientationParams: Orientation.
ParagraphInfo: Paragraph information.
RecognitionFailed: Recognition failed
RecognitionResults: Text recognition results.
Rectangle: Rectangle.
ResultIter: An iterator over recognition results.
Tesseract: Common tesseract methods.
Text: Tesseract-specific string.
TextElement: Text element.
TextRecognizer: OCR engine interface.
Utf8Text: Tesseract-specific UTF-8 string.
WriteFailed: Write failed

Enums§

BlockType: Block type.
LayoutLevel: Element of layout hierarchy
OcrEngineMode: OCR engine mode
Orientation: Text position on the page.
PageSegmentationMode: Page layout analysis modes.
ParagraphJustification: Paragraph alignment.
TextlineOrder: The order of the lines of text.
WritingDirection: Text writing direction.

Functions§

leptonica_version: Returns Leptonica library version.
version: Returns Tesseract library version.