OCR Result

Reference documentation on processing an OCR result using the Mindee SDKs.

You'll need to have a response object as described in the Response Processing section.

Accessing the Page OCR

An OCRPage describes the text and words of a single page in the document.

OCRPage Attributes

Content

Full text content extracted from the document page.

Words

List of all words found on the page.

Each word has the following properties:

• Content: Text content of the word.
• Polygon: Coordinates of the detected word.

Sample Code

Python

from mindee import OCRResponse

def handle_response(response: OCRResponse) -> None:
    pages = response.inference.result.pages

    for page in pages:
        # Page-level properties
        page_content = page.content
        words = page.words

        # Access the full text content extracted from this page.
        print(f"Page content: {page_content}")

        # Access all words detected on this page.
        for word in words:
            word_content = word.content
            word_polygon = word.polygon

            print(f"Word: {word_content}")
            print(f"Polygon: {word_polygon}")

Node.js

handleResponse(response) {
  const pages = response.inference.result.pages;

  for (const page of pages) {
    // Full text content extracted from this page.
    const pageContent = page.content;
    console.log(`Page content: ${pageContent}`);

    // All words detected on this page.
    for (const word of page.words) {
      const wordContent = word.content;
      const wordPolygon = word.polygon;

      console.log(`Word: ${wordContent}`);
      console.log(`Polygon: ${wordPolygon.toString()}`);
    }
  }
}

PHP

use Mindee\V2\Product\Ocr\OcrResponse;

public function handleResponse(OcrResponse $response)
{
    $pages = $response->inference->result->pages;
}

Ruby

def handle_response(response)
  pages = response.inference.result.pages

  pages.each do |page|
    # Page-level properties
    page_content = page.content
    words = page.words

    # Access the full text content extracted from this page.
    puts "Page content: #{page_content}"

    # Access all words detected on this page.
    words.each do |word|
      word_content = word.content
      word_polygon = word.polygon

      puts "Word: #{word_content}"
      puts "Polygon: #{word_polygon}"
    end
  end
end

Java

import com.mindee.v2.product.ocr.OcrPage;
import com.mindee.v2.product.ocr.OcrResponse;
import com.mindee.v2.product.ocr.OcrWord;

public void handleResponse(OcrResponse response) {
    var pages = response.getInference().getResult().getPages();

    for (OcrPage page : pages) {
        // Page-level properties
        String pageContent = page.getContent();
        var words = page.getWords();

        // Access the full text content extracted from this page.
        System.out.println("Page content: " + pageContent);

        // Access all words detected on this page.
        for (OcrWord word : words) {
            String wordContent = word.getContent();
            var wordPolygon = word.getPolygon();

            System.out.println("Word: " + wordContent);
            System.out.println("Polygon: " + wordPolygon.toString());
        }
    }
}

.NET

using Mindee.V2.Product.Ocr;

public void HandleResponse(OcrResponse response)
{
    var pages = response.Inference.Result.Pages;

    foreach (var page in pages)
    {
        // Page-level properties
        string pageContent = page.Content;
        var words = page.Words;

        // Access the full text content extracted from this page.
        Console.WriteLine($"Page content: {pageContent}");

        // Access all words detected on this page.
        foreach (var word in words)
        {
            string wordContent = word.Content;
            var wordPolygon = word.Polygon;

            Console.WriteLine($"Word: {wordContent}");
            Console.WriteLine($"Polygon: {wordPolygon}");
        }
    }
}