OCR support means the ability to connect an external module (library) for the recognition of printed text (optical character recognition, OCR) on images, either separate or embedded in documents.
To connect OCR, you need to implement the IOcrConnector interface in the client code and assign the connector to the OcrConnector property of the OcrIndexingOptions object of the indexing options. The Recognize method receives an OcrContext object, which provides the image stream, the location of the image (ImageLocation: a separate image file, an image embedded in a document or an image that is an item of a container) and the image file extension. The method returns the recognized text. The OcrIndexingOptions class also specifies for which images text recognition is enabled (see Indexing options).
The following example demonstrates how to implement the OCR connector using Aspose.OCR library for text recognition in images.
C#
stringindexFolder=@"c:\MyIndex";stringdocumentFolder=@"c:\MyDocuments";// Creating an indexIndexindex=newIndex(indexFolder);// Setting the OCR indexing optionsIndexingOptionsoptions=newIndexingOptions();options.OcrIndexingOptions.EnabledForSeparateImages=true;options.OcrIndexingOptions.EnabledForEmbeddedImages=true;options.OcrIndexingOptions.OcrConnector=newAsposeOcrConnector();// Indexing documents in a document folderindex.Add(documentFolder,options);// Searching in the indexSearchResultresult=index.Search("Einstein");...// Implementing the OCR connector that uses Aspose.OCR library// You need to install the following package:// https://www.nuget.org/packages/Aspose.OCR/publicclassAsposeOcrConnector:IOcrConnector{publicAsposeOcrConnector(){}publicstringRecognize(OcrContextcontext){stringextension=context.ImageFileExtension.ToLowerInvariant();if(context.ImageLocation==ImageLocation.Separate){switch(extension){case".gif":case".png":case".jpg":case".jpeg":case".bmp":case".tif":case".tiff":case"":returnRecognizePrivate(context);default:returnnull;}}elseif(context.ImageLocation==ImageLocation.Embedded||context.ImageLocation==ImageLocation.ContainerItem){returnRecognizePrivate(context);}else{thrownewNotSupportedException("The image type is not supported: "+context.ImageLocation);}}privatestringRecognizePrivate(OcrContextcontext){MemoryStreammemoryStream=newMemoryStream();context.ImageStream.CopyTo(memoryStream);memoryStream.Position=0;// Rewinding the stream before recognitionAsposeOcrasposeOcr=newAsposeOcr();stringresult=asposeOcr.RecognizeImage(memoryStream);returnresult;}}
The next example demonstrates how to implement the OCR connector using Aspose Cloud OCR API.
C#
// Implementing the OCR connector that uses Aspose Cloud OCR// The full API for Aspose Cloud OCR for .NET you can find in the repository:// https://github.com/aspose-ocr-cloud/aspose-ocr-cloud-dotnet// Sid and key you can get after free registration at// https://dashboard.aspose.cloud/applicationspublicclassAsposeCloudOcrConnector:IOcrConnector{privatereadonlyConfiguration_configuration;publicAsposeCloudOcrConnector(){_configuration=newConfiguration();_configuration.AppSid="...";_configuration.AppKey="...";}publicstringRecognize(OcrContextcontext){OcrApiapi=newOcrApi(_configuration);varrequest=newPostOcrFromUrlOrContentRequest(context.ImageStream);OCRResponseresponse=api.PostOcrFromUrlOrContent(request);returnresponse.Text;}}
And the following example demonstrates how to implement the OCR connector using Tesseract.
C#
// Implementing the OCR connector that uses Tesseract// You need to install the following packages:// https://www.nuget.org/packages/Tesseract/// https://www.nuget.org/packages/Tesseract.Data.English/publicclassTesseractOcrConnector:IOcrConnector{publicTesseractOcrConnector(){}publicstringRecognize(OcrContextcontext){byte[]buffer;using(MemoryStreammemoryStream=newMemoryStream()){context.ImageStream.CopyTo(memoryStream);// Reading the whole image streambuffer=memoryStream.ToArray();}// The language data are expected in the tessdata folder of the application folderstringpath=Path.Combine(AppContext.BaseDirectory,"tessdata");using(varengine=newTesseractEngine(path,"eng",EngineMode.Default))using(Piximg=Pix.LoadFromMemory(buffer))using(PagerecognizedPage=engine.Process(img)){stringrecognizedText=recognizedPage.GetText();returnrecognizedText;}}}
OCR in the extraction in a separate process
Since version 26.11, the text of the images of the documents that are extracted in a separate process can be recognized as well. The OCR connector of the indexing options is not passed to the extraction process: assign the connector to the ExtractionHost.OcrConnector property in the program of that process, and enable the recognition in the OcrIndexingOptions of the indexing options. When the options enable the recognition and the extraction process has no OCR connector, the images are not recognized, and the index reports it once per operation by the ErrorOccurred event. An example is presented on the Extraction in separate process page.
More resources
GitHub examples
You may easily run the code from documentation articles and see the features in action in our GitHub examples: