如何在 C# 中使用 OCR 车牌（教程）

To get started with IronOCR in your C# project, you’ll need to install the IronOCR package from NuGet. IronOCR is compatible with .NET Framework and .NET Core, making it easy to integrate into various .NET applications.

Step 1: Open Package Manager Console

In Visual Studio, navigate to Tools > NuGet Package Manager > Package Manager Console.

Step 2: Install the IronOCR Package

Enter the following command in the Package Manager Console:

Install-Package IronOcr

This command installs the IronOCR library, including all dependencies needed to run OCR functions in your project. As our application requires advanced features like license plate detection using computer vision, you can also install the optional IronOcr.ComputerVision.Windows package this way:

Install-Package IronOcr.ComputerVision.Windows

Ensure you have the IronOcr.Extensions.AdvancedScan extension installed so you can utilize its powerful ReadLicensePlate method:

Install-Package IronOcr.Extensions.AdvancedScan

Alternatively, you can install the packages using Tools > NuGet Package Manager > Manage NuGet Packages for Solution and search for the packages you need:

Setting Up the Code

Finally, we must add the necessary imports and using statements to the top of our code:

using IronOcr;

using IronOcr;

Imports IronOcr

In this section, we will create a program to read license plates using IronOCR, a Tesseract OCR Engine that excels at extracting text from images. To implement vehicle detection, we may also incorporate additional machine-learning libraries. Notably, IronOCR integrates with OpenCV, a leading open-source computer vision library, allowing us to perform object detection tasks like identifying vehicles and license plates.

Example License Plate Image

We'll be working with the following license plate:

Example Code to Perform OCR on License Plate

using IronOcr;

var ocr = new IronTesseract();

using (var input = new OcrImageInput("licensePlate.jpeg"))
{
    // Fixes digital noise and makes the image easier to read
    input.DeNoise();
    input.ToGrayScale();

    // Reads the license plate information and stores it for further use
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);

    // Saves the license plate text to a string variable
    string output = result.Text;

    // Outputs the license plate text to the console
    Console.WriteLine(output);
}

using IronOcr;

var ocr = new IronTesseract();

using (var input = new OcrImageInput("licensePlate.jpeg"))
{
    // Fixes digital noise and makes the image easier to read
    input.DeNoise();
    input.ToGrayScale();

    // Reads the license plate information and stores it for further use
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);

    // Saves the license plate text to a string variable
    string output = result.Text;

    // Outputs the license plate text to the console
    Console.WriteLine(output);
}

Imports IronOcr

Private ocr = New IronTesseract()

Using input = New OcrImageInput("licensePlate.jpeg")
	' Fixes digital noise and makes the image easier to read
	input.DeNoise()
	input.ToGrayScale()

	' Reads the license plate information and stores it for further use
	Dim result As OcrLicensePlateResult = ocr.ReadLicensePlate(input)

	' Saves the license plate text to a string variable
	Dim output As String = result.Text

	' Outputs the license plate text to the console
	Console.WriteLine(output)
End Using

Code Breakdown:

• Initialization: var ocr = new IronTesseract(); creates a new instance of the IronTesseract class, which provides methods for Optical Character Recognition (OCR).

• Image Input: The using statement creates a new OcrImageInput object with the specified image file "licensePlate.jpeg". This object is designed to hold the image data for OCR processing.

• Image Preprocessing:

  • input.DeNoise(); applies a digital noise reduction filter to enhance the quality of the image, making it easier for the OCR engine to read the text.
  • input.ToGrayScale(); converts the image to grayscale, which can improve recognition accuracy and processing speed.

• License Plate Recognition: The line OcrLicensePlateResult result = ocr.ReadLicensePlate(input); uses the ReadLicensePlate method to analyze the processed image and extract any license plate information it detects, storing the results in an OcrLicensePlateResult object.

• Output Storage: The license plate text is stored in the string variable output by accessing result.Text, which contains the recognized text from the license plate.

• Console Output: Finally, Console.WriteLine(output); prints the extracted license plate text to the console for verification.

If we have an image of an entire car rather than just the license plate, we can specify a rectangular region to focus on the license plate area. We can use System.Drawing.Rectangle to define this area in pixels.

Original Image

We’ll use the following image file for our example:

By specifying the area of interest, we improve processing speed and avoid extracting unnecessary text.

Implementation Code

using IronOcr;
using System.Drawing;

var ocr = new IronTesseract();

using (var input = new OcrInput())
{
    var contentArea = new Rectangle(x: 252, y: 282, width: 148, height: 47);
    input.LoadImage("CarPlate.jpeg", contentArea);
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);
    Console.WriteLine(result.Text);
}

using IronOcr;
using System.Drawing;

var ocr = new IronTesseract();

using (var input = new OcrInput())
{
    var contentArea = new Rectangle(x: 252, y: 282, width: 148, height: 47);
    input.LoadImage("CarPlate.jpeg", contentArea);
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);
    Console.WriteLine(result.Text);
}

Imports IronOcr
Imports System.Drawing

Private ocr = New IronTesseract()

Using input = New OcrInput()
	Dim contentArea = New Rectangle(x:= 252, y:= 282, width:= 148, height:= 47)
	input.LoadImage("CarPlate.jpeg", contentArea)
	Dim result As OcrLicensePlateResult = ocr.ReadLicensePlate(input)
	Console.WriteLine(result.Text)
End Using

Code Breakdown:

• Initialization: var ocr = new IronTesseract(); creates a new instance of the IronTesseract class, which is responsible for executing the OCR operations.
• OCR Input: The using (var input = new OcrInput()) statement creates a new OcrInput object, which will be used to load and process the image for OCR.
• Defining the Area of Interest: var contentArea = new Rectangle(x: 252, y: 282, width: 148, height: 47); defines a rectangular area (contentArea) within the image. This rectangle specifies the coordinates and dimensions (width and height) where the license plate is expected to be located.
• Loading the Image: input.LoadImage("CarPlate.jpeg", contentArea); loads the specified image file ("CarPlate.jpeg") and focuses on the defined rectangle (contentArea) to limit the OCR processing to that specific region.
• Reading the License Plate: OcrLicensePlateResult result = ocr.ReadLicensePlate(input); invokes the ReadLicensePlate method, which analyzes the input image for license plate characters and returns an OcrLicensePlateResult object containing the extracted text.
• Output: Console.WriteLine(result.Text); prints the recognized text from the license plate to the console.

IronOCR leverages OpenCV to identify text regions within images, employing various image processing techniques. This functionality enables programs to detect license plates by locating text areas in the image and then utilizing Tesseract to read those regions.

Installation

To enable the license plate detection model, install the required package via the Package Manager Console:

Example using automatic region detection for license plates:

var ocr = new IronTesseract();

using (var input = new OcrImageInput("CarPlate.jpeg"))
{
    input.FindTextRegion();
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);
    Console.WriteLine(result.Text);
}

var ocr = new IronTesseract();

using (var input = new OcrImageInput("CarPlate.jpeg"))
{
    input.FindTextRegion();
    OcrLicensePlateResult result = ocr.ReadLicensePlate(input);
    Console.WriteLine(result.Text);
}

Dim ocr = New IronTesseract()

Using input = New OcrImageInput("CarPlate.jpeg")
	input.FindTextRegion()
	Dim result As OcrLicensePlateResult = ocr.ReadLicensePlate(input)
	Console.WriteLine(result.Text)
End Using

Code Breakdown:

• Initialization: An instance of the IronTesseract class is created, which will be used for optical character recognition (OCR) with the Tesseract engine.
• Image Input: A new OcrImageInput object is instantiated with the specified image file ("CarPlate.jpeg"). This object will serve as the input for the OCR process and is enclosed in a using statement to ensure proper resource management.
• Text Region Detection: The FindTextRegion() method is called on the input object. This method employs computer vision techniques to automatically identify areas within the image that likely contain text, specifically targeting the license plate.
• License Plate Recognition: The ReadLicensePlate method is invoked to analyze the detected text regions and extract the license plate number. The result is stored in an OcrLicensePlateResult object, which holds the recognized text and any associated metadata.
• Output: The detected license plate text is printed to the console, allowing the user to see the extracted license plate number.