Tesseract

Namespace: Wisej.Ext.Tesseract

Assembly: Wisej.Ext.Tesseract (4.1.0.0)

• Component
  • Component
    • Tesseract

A component for extending the functionality of the Wisej Camera to include text detection.

C#

public class Tesseract : Component

VB.NET

Public Class Tesseract
    Inherits Component

See: https://github.com/naptha/tesseract.js#tesseractjs.

Constructors

Tesseract()

Initializes a new instance of the Tesseract without a specified container.

Example:

// Create the component in code and attach it to an existing camera.
var tesseract = new Tesseract
{
	Camera = this.camera1,
	Language = "eng",
	Interval = 1000,
	MinimumConfidence = 70
};

tesseract.TextRecognized += (s, e) => label1.Text = e.Text;

Tesseract(container)

Initializes a new instance of the Tesseract class with a specified container.

Name	Type	Description
container	IContainer	An IContainer that represents the container of the component.

Throws:

• ArgumentNullException container is null.

Example:

// Typically generated by the designer in InitializeComponent(),
// so the component is disposed together with its container.
this.components = new System.ComponentModel.Container();
this.tesseract1 = new Wisej.Ext.Tesseract.Tesseract(this.components);
this.tesseract1.Camera = this.camera1;

Properties

Camera

Control: Returns or sets the Wisej.NET Camera instance to attach to. (Default: null)

See Wisej.Web.Ext.Camera

Enabled

Boolean: Returns or sets whether text-detection is enabled for the associated Camera. Set to true by default. (Default: True)

Interval

Int32: Returns or sets the interval to use in between text detection requests in the live camera preview. (Default: 1500)

The interval is specified in milliseconds.

Keywords

String[]: Returns or sets a list of keywords to filter for. (Default: null)

Language

String: Returns or sets the detection language to use. (Default: "eng")

See: https://tesseract-ocr.github.io/tessdoc/Data-Files#data-files-for-version-400-november-29-2016

MinimumConfidence

Int32: Returns or sets the minimum confidence interval to accept.

ShowWords

Boolean: Returns or sets whether to show bounding rectangles around detected words in the live camera preview. (Default: True)

Defaults to true.

Whitelist

String: Returns or sets the white listed characters to scan for. (Default: null)

Setting white list characters makes the result only contain these characters, useful if content in image is limited.

WorkerCount

Int32: Returns or sets the number of workers used to detect text in the live camera preview. (Default: 1)

A Tesseract.js worker is an object that creates and manages an instance of Tesseract. Think of a worker as a thread that is responsible for doing OCR (Optical Character Recognition) jobs. A higher number of workers is better, but results in increased CPU and battery consumption.

Methods

OnAddReferences(list)

Ensures the Camera widget exists in the ComponentManager.

Parameter	Type	Description
list	IList

OnWebEvent(e)

Fires events from the client.

Parameter	Type	Description
e	WisejEventArgs	WisejEventArgs

OnWebRender(config)

Provides the client widget's configuration.

Parameter	Type	Description
config	Object	The configuration.

ScanImageAsync(imageSource)

Scans a given image source (url) and return the discovered text.

Parameter	Type	Description
imageSource	String	The input image url.

Returns: Task<ScanResult>. ScanResult object containing the text, list of words, and confidence (accuracy) of the result.

The image source can be an absolute or relative URL, a Wisej.NET resource URL (e.g. the ImageSource of a PictureBox pointing to an embedded resource), or a base64 data URL. To scan an Image instance, use the ScanImageAsync overload instead.

Example:

// Scan an image from a website.
ScanResult result = await tesseract1.ScanImageAsync("https://www.example.com/images/sample.png");
AlertBox.Show(result.Text);

// Scan an embedded resource displayed in a PictureBox.
// Set the build action of the image to "Embedded Resource" and
// assign its resource path to the ImageSource property.
pictureBox1.ImageSource = "resource.wx/WisejWebPageApplication/Images/ITGlogo.png";
result = await tesseract1.ScanImageAsync(pictureBox1.ImageSource);

if (result.Confidence >= tesseract1.MinimumConfidence)
	AlertBox.Show(string.Join(", ", result.Words));

ScanImageAsync(image)

Scans a given Image and return the discovered text.

Parameter	Type	Description
image	Image	The image to be scanned Image.

Returns: Task<ScanResult>. ScanResult object containing the text, list of words, and confidence (accuracy) of the result.

The image is converted to a PNG base64 data URL and passed to ScanImageAsync. The image can come from any source, such as the Image of a PictureBox set in the designer, a file, or an embedded resource.

Example:

// Scan the image assigned to a PictureBox in the designer.
ScanResult result = await tesseract1.ScanImageAsync(pictureBox1.Image);
AlertBox.Show(result.Text);

// Load an embedded resource ending in "YourImage.png" and scan it.
System.Reflection.Assembly assembly = this.GetType().Assembly;
string resourceName = assembly.GetManifestResourceNames()
	.FirstOrDefault(name => name.EndsWith("YourImage.png"));

if (!String.IsNullOrEmpty(resourceName))
{
	using (System.IO.Stream resource = assembly.GetManifestResourceStream(resourceName))
	using (Image image = Image.FromStream(resource))
	{
		result = await tesseract1.ScanImageAsync(image);
		AlertBox.Show($"{result.Text} (confidence: {result.Confidence})");
	}
}

Events

TextRecognized

EventHandler<TextRecognizedEventArgs> Fired when text is recognized by the camera.

TextRecognizedEventArgs contains information about the text discovered, the words in the text, and the confidence (accuracy) of the recognition.

WorkerError

EventHandler<String> Fired when a worker throws an error.

A Tesseract.js worker is an object that creates and manages an instance of Tesseract. Think of a worker as a thread that is responsible for doing OCR (Optical Character Recognition) jobs.