How to Get Text From Invoice in C# Tutorial

How to OCR Receipt in Tesseract

1. Install C# library OCR receipt with Tesseract
2. Explore features rich C# library for performing OCR on receipt
3. Extract data from receipt with Tesseract
4. Search in the extracted text result for specific data
5. Read barcodes value on the input receipt image

1. IronOCR, An Optical Character Recognition API

IronOCR is an OCR library that can be used to recognize text data from images for information extraction, including receipt OCR. It is built on the Tesseract OCR engine, which is considered one of the most accurate OCR engines available to date for receipt recognition. IronOCR can read key information from different document types, including PNG, JPG, TIFF, JSON, and PDF formats, and it can recognize text in multiple languages.

One of the key features of IronOCR that makes it particularly useful for receipt OCR is its ability to automatically detect text orientation, even if the image has been rotated or skewed. This is essential for accurate text recognition on receipts uploads and data extraction, as receipts often contain a lot of information and can be folded or crumpled, causing the text to become skewed.

2. IronOCR Features

• C# OCR uses Deep Learning to scan and recognize texts from pictures, scanned documents, and PDFs.
• .NET OCR supports more than 127 global languages.
• IronOCR can read text from images in many file formats, including PNG, JPG, TIFF, and PDF.
• Text, structured data, JSON output, or searchable PDFs can be produced from extracted information.
• IronOCR supports .NET versions 5, 6, and 7 (Core, Framework, and Standard).
• IronOCR divides the input into different pictures based on text regions. It uses Computer Vision to identify areas that contain text elements.

3. Creating a New Project in Visual Studio

Open Visual Studio and go to the File menu. Select "New Project" and then choose Console Application.

Enter the project name and select the path in the appropriate text box. Then, click the Create button. Select the required .NET Framework, as shown in the screenshot below:

The project structure for the Console Application will now be generated. Once finished, it will open the Program.cs file, in which you can write and execute source code.

4. Install IronOCR

In Visual Studio, you can integrate IronOCR with your C# project easily. IronOCR offers multiple ways to integrate with a C# .NET project. Here, we'll discuss one of them: Installing IronOCR using the NuGet Package Manager.

In Visual Studio, go to Tools > NuGet Package Manager > Package Manager Console

A new console will appear at the bottom of Visual Studio's window. Type the below command in the console and press enter.

Install-Package IronOcr

IronOCR will be installed in just a few seconds.

5. Data Extraction from Receipts Using IronOCR

IronOCR is a powerful OCR library that can be used to extract and access detailed data from receipts. With IronOCR, you can convert a picture of a receipt into machine-readable text that can be easily analyzed and processed without compromising data privacy.

Here's an example of how you can use IronOCR to extract text from a receipt:

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        IronTesseract ocrTesseract = new IronTesseract();

        // Load the receipt image
        using (OcrInput ocrInput = new OcrInput("ocr.png"))
        {
            // Read the OCR result
            OcrResult ocrResult = ocrTesseract.Read(ocrInput);
            string recognizedText = ocrResult.Text;

            // Output the recognized text to the console
            Console.WriteLine(recognizedText);
        }
    }
}

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        IronTesseract ocrTesseract = new IronTesseract();

        // Load the receipt image
        using (OcrInput ocrInput = new OcrInput("ocr.png"))
        {
            // Read the OCR result
            OcrResult ocrResult = ocrTesseract.Read(ocrInput);
            string recognizedText = ocrResult.Text;

            // Output the recognized text to the console
            Console.WriteLine(recognizedText);
        }
    }
}

Imports IronOcr
Imports System

Friend Class Program
	Shared Sub Main()
		Dim ocrTesseract As New IronTesseract()

		' Load the receipt image
		Using ocrInput As New OcrInput("ocr.png")
			' Read the OCR result
			Dim ocrResult As OcrResult = ocrTesseract.Read(ocrInput)
			Dim recognizedText As String = ocrResult.Text

			' Output the recognized text to the console
			Console.WriteLine(recognizedText)
		End Using
	End Sub
End Class

Refer to the Reading Text from Image tutorial for further details on how IronOCR reads text from images using C#.

The output of the code above:

- LOGO SHOP
- LOREM IPSUM
- DOLOR SIT AMET CONSECTETUR
- ADIPISCING ELIT
- 1 LOREM IPSUM $3.20
- 2 ORNARE MALESUADA $9.50
- 3 PORTA FERMENTUM $5.90
- 4 SODALES ARCU $6.00
- 5 ELEIFEND $9.00
- 6 SEM NISIMASSA $0.50
- 7 DUIS FAMES DIS $7.60
- 8 FACILISI RISUS $810
- TOTAL AMOUNT $49.80
- CASH $50.00

6. Specific Data Extraction From Receipt Image Using IronOCR

IronOCR allows developers to retrieve crucial information from scanned receipts, such as tax amounts and merchant names.

Here is an example demonstrating how to extract the total amount value from a receipt image:

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        IronTesseract ocrTesseract = new IronTesseract();

        // Set the language for OCR
        ocrTesseract.Language = OcrLanguage.English;

        // Load the receipt image
        using (OcrInput ocrInput = new OcrInput("ocr.png"))
        {
            // Optimize the input image for OCR
            ocrInput.DeNoise(true);
            ocrInput.Contrast();
            ocrInput.EnhanceResolution();
            ocrInput.ToGrayScale();

            OcrResult ocrResult = ocrTesseract.Read(ocrInput);

            // Search for the total amount in the OCR result
            var totalAmount = ocrResult.Text.Contains("Total:") ? ocrResult.Text.Split("Total:")[1].Split("\n")[0] : "";
            Console.WriteLine("Total Amount: " + totalAmount);
        }
    }
}

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        IronTesseract ocrTesseract = new IronTesseract();

        // Set the language for OCR
        ocrTesseract.Language = OcrLanguage.English;

        // Load the receipt image
        using (OcrInput ocrInput = new OcrInput("ocr.png"))
        {
            // Optimize the input image for OCR
            ocrInput.DeNoise(true);
            ocrInput.Contrast();
            ocrInput.EnhanceResolution();
            ocrInput.ToGrayScale();

            OcrResult ocrResult = ocrTesseract.Read(ocrInput);

            // Search for the total amount in the OCR result
            var totalAmount = ocrResult.Text.Contains("Total:") ? ocrResult.Text.Split("Total:")[1].Split("\n")[0] : "";
            Console.WriteLine("Total Amount: " + totalAmount);
        }
    }
}

Imports Microsoft.VisualBasic
Imports IronOcr
Imports System

Friend Class Program
	Shared Sub Main()
		Dim ocrTesseract As New IronTesseract()

		' Set the language for OCR
		ocrTesseract.Language = OcrLanguage.English

		' Load the receipt image
		Using ocrInput As New OcrInput("ocr.png")
			' Optimize the input image for OCR
			ocrInput.DeNoise(True)
			ocrInput.Contrast()
			ocrInput.EnhanceResolution()
			ocrInput.ToGrayScale()

			Dim ocrResult As OcrResult = ocrTesseract.Read(ocrInput)

			' Search for the total amount in the OCR result
			Dim totalAmount = If(ocrResult.Text.Contains("Total:"), ocrResult.Text.Split("Total:")(1).Split(vbLf)(0), "")
			Console.WriteLine("Total Amount: " & totalAmount)
		End Using
	End Sub
End Class

Thanks to the multiple settings provided by the OcrInput class, it is possible to optimize the input image for better accuracy in the OCR process.

Input

Output

- Total 16.5

7. Read Barcodes on Receipts

IronOCR can be used to read barcodes on receipts as well as text. To read barcodes on receipts, you will need to use the BarcodeReader class in combination with the ReadBarCodes method.

Here's an example of how to read barcodes:

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        var ocrTesseract = new IronTesseract();
        ocrTesseract.Configuration.ReadBarCodes = true;

        // Load the receipt image with a barcode
        using (var ocrInput = new OcrInput("b.png"))
        {
            OcrResult ocrResult = ocrTesseract.Read(ocrInput);

            // Output the barcode values to the console
            foreach (var barcode in ocrResult.Barcodes)
            {
                Console.WriteLine(barcode.Value);
            }
        }
    }
}

using IronOcr;
using System;

class Program
{
    static void Main()
    {
        var ocrTesseract = new IronTesseract();
        ocrTesseract.Configuration.ReadBarCodes = true;

        // Load the receipt image with a barcode
        using (var ocrInput = new OcrInput("b.png"))
        {
            OcrResult ocrResult = ocrTesseract.Read(ocrInput);

            // Output the barcode values to the console
            foreach (var barcode in ocrResult.Barcodes)
            {
                Console.WriteLine(barcode.Value);
            }
        }
    }
}

Imports IronOcr
Imports System

Friend Class Program
	Shared Sub Main()
		Dim ocrTesseract = New IronTesseract()
		ocrTesseract.Configuration.ReadBarCodes = True

		' Load the receipt image with a barcode
		Using ocrInput As New OcrInput("b.png")
			Dim ocrResult As OcrResult = ocrTesseract.Read(ocrInput)

			' Output the barcode values to the console
			For Each barcode In ocrResult.Barcodes
				Console.WriteLine(barcode.Value)
			Next barcode
		End Using
	End Sub
End Class

Input Image

Output Text

8. Conclusion

The article above explains the process of installing and using IronOCR in a C# project to extract data from receipts, with example code snippets provided.

Please read the tutorial on reading text from images.

IronOCR is a part of the Iron Suite, which includes five different .NET libraries for manipulating documents and images. You can buy the entire Iron Suite for the price of just two IronOCR licenses.

Try IronOCR in your production apps with a free trial.