OCR Documents in .NET
OCR Documents in .NET.
Overview
Dynamsoft’s OCR SDK is an add-on of Dynamic .NET TWAIN, an image acquisition SDK optimized for .NET applications. The OCR SDK allows you to convert scanned images to searchable PDF/text files. Recognized as a useful feature, it’s not easy to implement it. A lot of complicated things, such as accuracy, image format and more, are involved to get better results. OCR performance is another important factor that affects the efficiency of the whole process.
Dynamsoft’s OCR SDK, optimized based on the highly developed open source engine (Tesseract OCR engine), helps you relieve from these burdens. By integrating with Dynamic .NET TWAIN, you can create a robust image acquisition and processing solution in several lines of source code.
Key Features
- Supports more than 40 languages, including Arabic and various Asian languages.
- High OCR performance by supporting multi-thread processing.
- Accurate recognition with font identification
- Easy integration with the image acquisition SDK – Dynamic .NET TWAIN.
The following sections will show you how to integrate the OCR add-on to your WinForm application and convert scanned images to searchable PDF/text files.
Source Code
1. Embed Dynamic .NET TWAIN to your WinForm or WPF app.
We will take WinForm as an example.
Assume you’ve already downloaded and installed the .NET component onto your development machine (If not, please download the 30-day free trial from Dynamsoft’s website.).
Open your WinForm app or create a new one in Visual Studio. From the Tools menu, select Choose Toolbox Items. In the prompt dialog box, click Browse and select DynamicDotNetTWAIN.dll which can be found in the installation folder of Dynamic .NET TWAIN. Click OK to close the dialog box.
Drag and drop the component to the form.
2. Scan images from scanners, webcams or get from local folders.
Dynamic .NET TWAIN supports getting images from various sources, including scanners, webcams and other TWAIN/WIA/UVC compatible devices. In this article, I’ll show you how to load an existing image from your local disk.
SetViewMode: defines the view mode of the control.
LoadImage: loads the existing local images. Supported image format
includes BMP, PNG, JPEG, TIFF (both single and multi-page) and PDF (both single
and multi-page).
this.dynamicDotNetTwain1.SetViewMode(1, 1);
OpenFileDialog filedlg = new OpenFileDialog();
if (filedlg.ShowDialog() == DialogResult.OK)
{
foreach (string strfilename in filedlg.FileNames)
{
this.dynamicDotNetTwain1.LoadImage(strfilename);
}
}
3. Initialize the OCR add-on and choose the language package.
1) Choose
the language package and define the path of the package by using the OCRTessDataPath
property.
Dynamsoft’s OCR SDK supports more than 40 languages, including English, Spanish, Arabic and more. The sample code below chooses English as the default language. Other language packages can be downloaded from Dynamsoft’s website: OCR SDK Language Packages
string languageFolder = Application.StartupPath;
this.dynamicDotNetTwain1.OCRTessDataPath = languageFolder;
this.dynamicDotNetTwain1.OCRLanguage = "eng";
2) Set the path of DynamicOCR.dll or DynamicOCRx64.dll to initialize the OCR add-on.
this.dynamicDotNetTwain1.OCRDllPath = "";
3) Choose the OCR result file format and save. Supported file format includes Text, PDF Plain Text and PDF Image over Text. By setting the format to PDF Image over Text, the detailed image/text position and format, such as font names, font sizes, line widths and more, will keep as original.
this.dynamicDotNetTwain1.OCRResultFormat = (Dynamsoft.DotNet.TWAIN.OCR.ResultFormat)this.ddlResultFormat.SelectedIndex;
byte[] sbytes = this.dynamicDotNetTwain1.OCR(this.dynamicDotNetTwain1.CurrentSelectedImageIndicesInBuffer);
if(sbytes != null)
{
SaveFileDialog filedlg = new SaveFileDialog();
if (this.ddlResultFormat.SelectedIndex != 0)
{
filedlg.Filter = "PDF File(*.pdf)| *.pdf";
}
else
{
filedlg.Filter = "Text File(*.txt)| *.txt";
}
if (filedlg.ShowDialog() == DialogResult.OK)
{
FileStream fs = File.OpenWrite(filedlg.FileName);
fs.Write(sbytes, 0, sbytes.Length);
fs.Close();
}
}
else
{
MessageBox.Show(this.dynamicDotNetTwain1.ErrorString);
}
Distribution
To distribute the application to the end users, please copy the following files to the client machine along with the EXE file.
The
language package
DynamicOCR.dll (for 32-bit Windows OS) and/or DynamicOCRx64.dll (for 64-bit
Windows OS)
DynamicDotNetTwain.dll
Xcopy deployment is also supported.
Resources
The complete source code of OCR can be downloaded from the article. To test and/or customize the code, you can download the trial version of Dynamic .NET TWAIN from Dynamsoft’s website.
Download Dynamic .NET TWAIN 30-Day Free Trial
Other demos/samples of .NET image acquisition and processing can be found here:
If you have any questions, you can contact our support team at nettwain@dynamsoft.com.