What's New in ABBYY FineReader Engine 12
Here you can find the list of new features in ABBYY FineReader Engine 12.
New and improved language recognition
- Bangla language recognition
- Arabic_Amount special language recognition
See the list of predefined languages in ABBYY FineReader Engine.
OCR and ICR improvements
- Synthesis of business cards can now be performed with the usage of additional parameters (IFRPage::SynthesizeBusinessCardEx, ISynthesisParamsForPage::BusinessCardSynthesisParams)
- You may now use the new Accurate recognition mode (IRecognizerParams::Mode)
- New property to check the presence of the barcodes except for the postal ones (IBarcodeParams::EnableBarcodesCheck)
- You can now detect the barcode type, excluding the postal ones with the constant BT_AutodetectWithoutPostal (BarcodeTypeEnum)
- ABBYY FineReader Engine now recognizes Japanese Postal Customer Codes (BarcodeTypeEnum, Barcode Types)
Image opening and preprocessing
- You may now load the documents which were saved by the previously supported versions of ABBYY FineReader Engine (IFRDocument::ConvertFromOldVersion)
- You can obtain information about the presence of a digital signature in a document (IFRDocument::SourceHasDigitalSignature, IFRPage::SourceHasDigitalSignature)
- The text layer can now be checked for reliability (IFRDocument::CheckTextLayer)
- Improved default mode for processing digitally-born PDF files (SourceContentReuseModeEnum::CRM_Auto)
- The appearance of the output PDF file can now be changed (ITextLayerInjectionParams::AllowChangePDFAView)
- You may now determine and open files of the PDF Portfolio type (IEngine::IsPdfPortfolio, IEngine::IsPdfPortfolioFromStream, IFRDocument::AddImageFileFromAttachment)
Export
- Export to ALTO now supports ALTO versions 4.0, 4.1, 4.2 (AltoVersionEnum)
- New modes of fonts embedding into output PDF files (FontEmbeddingModeEnum)
- New interfaces to add the user-defined pictures to the output PDFs (IPDFPictures, IPDFPicture)
Code samples
- New Hello sample for .NET Core
Other improvements
- You can obtain information about the memory size allocated for the document object (IFRDocument::AllocatedSize)
- New options for releasing an Online License (IEngine::ReleaseOnlineLicense, "Release Online Licenses..." in the License Manager utility)
- The new "EnableIKeyLicenses" tag to use iKey hardware protection keys (Working with the LicensingSettings.xml File)
- ABBYY FineReader Engine installer now uses Windows Installer XML Toolset (WiX)
- New LICENSEDATADIR option to change the folder used for storing the auxiliary information (Installing the ABBYY FineReader Engine Library in Automatic Mode)
- ABBYY FineReader Engine 12 can now be run inside Azure Services
- You may now get a renewal date of the license limitations in the number of pages processed (ILicense::VolumeRefreshingDate)
Special features
- You may compare either the two documents or the two pages (IComparator::CompareDocuments, IComparator::ComparePages)
- The methods of Document Comparison API help to detect the changes in texts (IChange::GetAsTextChange) and export the comparison result to an XML or DOCX format (IComparisonResult::Export)
- You may upgrade your code using ABBYY ScanDifFinder SDK to a Document Comparison API implemented in ABBYY FineReader Engine (Migration Guide from ScanDifFinder SDK to Document Comparison API)
OCR and ICR improvements
- New supported postal barcodes: KIX, Royal Mail 4-State, Australian Post 4-State (BarcodeTypeEnum, Barcode Types)
Image opening and preprocessing
- Document layout can be saved and read from the stream (ILayout::SaveToStream, IEngine::CreateLayoutFromStream) or saved as an array of bytes (ILayout::SaveToArray)
Export
- Export to PDF supports the setting of the page orientation (IPDFExportFeatures::PageOrientation)
- Export to RTF supports tuning page margins (IRTFExportParams::PageMargins) and increasing page size if the contents do not fit on the page (IRTFExportParams::IncreasePaperSizeToFitContent)
- Export to XLSX supports picture embedding in the output document (IXLExportParams::WritePictures, IXLExportParams::PictureExportParams)
Code Samples
- New Document Comparison demo tool. This tool demonstrates how you can compare a document or page with its copy
- Updated CommandLineInterface sample
Other improvements
- ABBYY FineReader Engine HTML-help is now available in Javadoc format. It is included in the Java class library, so you can get the information about ABBYY FineReader Engine API from the IDEs that support Javadoc
- An Online License limits the number of CPU Cores available for one workstation
- An Online License now requires the usage of GoDaddy Trusted Root Certification Authority on the computer with the Licensing Service installed
This is the release for Windows and Linux versions of ABBYY FineReader Engine. All features from this release were included in Release 4 for the Mac OS version.
Special features
- MRZ can now be extracted from the document images (IFRPage::ExtractMrz)
- The data extracted from MRZ contains fields and lines (MrzField, MrzData, MrzLine) with information about the document and its holder and may be exported to an XML file or JSON file
New and improved language recognition
- Georgian language recognition
- Mathematical formula language recognition
See the list of predefined languages in ABBYY FineReader Engine.
Image opening and preprocessing
- You can now open digitally-born documents using the same methods as for opening images; the digital documents will be converted and their contents loaded (OfficeConversionSettings, OfficeConverters)
- You may set the binarization threshold during the image preprocessing (IPrepareImageMode::BinarizationThreshold)
- The mode of working with document pages in memory can be set before conversion of input images (IBatchProcessor::PageFlushingPolicy)
Export
- Creation and modification dates can now be viewed and changed. Also you may set up the dates' writing mode into the output document (DocumentContentInfo::CreationDate, DocumentContentInfo::ModificationDate, DocumentContentInfoWritingParams::WriteCreationDate, DocumentContentInfoWritingParams::WriteModificationDate). Only the dates in correct format can be written into the document. If during export an error is occurred, specify the date in a correct format or change its writing mode and call the export method again (Handling Errors During PDF/XPS Export)
- Original time format in the output document (TimeZoneTypeEnum)
Code samples
- New MRZExtraction sample. This sample demonstrates how to extract data from a machine-readable zone (MRZ)
- New code sample illustrating how to implement a timer that will abort processing if a method call is taking too long
Other improvements
- The blocks of data can now be sorted by their position (ILayout::SortedBlocks)
- An Online License does not require any specific settings, but you must set its parameters if you want to use the license with the proxy server
- An Online License is now allowed for the limited connections which number may be changed after autonomous run time of the license is over
- ABBYY FineReader Engine 12 can now be run inside Docker container
- The list of supported Java Development Kits is presented in System Requirements
- The number of ABBYY FineReader Engine build is now available before its modules loading. You may find it in the name of the JAR file which is included in ABBYY FineReader Engine Distribution Kit
- The file of End User License Agreement is now available in ABBYY FineReader Engine Distribution Kit
Special features
Improved Classification API:
- Updated classification technologies support classification on the basis of the image features, characteristics of the recognized text, or the combination of image and text characteristics (ClassifierTypeEnum)
- Classification is now available both for documents and for pages (IClassificationEngine::CreateObjectFromDocument, IClassificationEngine::CreateObjectFromPage)
- New classification models can be trained using k-fold cross-validation technique (ValidationParams), with detailed information about the model's performance (ValidationResult)
New and improved language recognition
- Burmese language recognition
- Improved Japanese (Modern) language for better recognition of Japanese texts which may include some Latin or Greek letters. It is now available for Business Card Recognition
- Updated Farsi OCR
See the list of predefined languages in ABBYY FineReader Engine.
OCR and ICR improvements
- Coordinates are available for all character recognition variants (ISynthesisParamsForPage::SaveCharacterCoordinatesForAllVariants)
Image opening and preprocessing
- Images can now be automatically cropped to document edges during preprocessing (IPagePreprocessingParams::CropImage)
- Noise reduction filter can be applied during preprocessing to improve the quality of recognition (IPagePreprocessingParams::ApplySigmaFilter, IImageDocument::ApplySigmaFilter)
- Anchor black squares can be found on the image (IImageDocument::FindBlackSquares)
Export
- PDF version 2.0 is supported for export
- Text layer injection supports tagged PDF files and compliance with the PDF/A standard
- PDF export supports AES-256 encryption in addition to RC4 , PDF/A-2b and PDF/A-3b conformance (IPDFExportParams::PDFAComplianceMode)
- Full support for ZUGFeRD-compliant PDF: documents exported with IPDFExportFeatures::EnableZugferdCompliance set to TRUE now pass the official ZUGFeRD validator
- Export to Excel supports right-to-left worksheet direction (IXLExportParams::CreateRtlSheets)
- Export to ALTO now supports ALTO version 3.1 (IAltoExportParams::AltoVersion)
- RTF export allows to retain more separators in output by disabling the separator filtration (IRTFExportParams::FilterSeparators)
- Better support for DOCX dotted separators
- HTML 5.0 support (HFM_Format50 constant of the HTMLFormatModeEnum)
- Support for PDF/UA-compliant documents (IPDFExportFeatures::EnablePDFUACompliance)
- Advanced layout emulation modes for export to TXT (ITextExportParams::LayoutRetentionMode)
Scanning API
- Paper size and scanning color mode can now be determined automatically (ScanPaperSizeEnum::SPS_Auto, ScanColorModeEnum::SCM_Auto)
- Empty pages can be skipped and image skew can be corrected automatically during scanning, if these features are supported by the scanner (IScanSourceSettings::DiscardBlankPages, IScanSourceSettings::CorrectSkew)
Code samples
- New code sample demonstrating Classification of the documents of any types
- Updated EnginesPool sample for Java
Other improvements
- Easier way to use FineReader Engine in multi-threaded Java applications (EnginesPool implementation added to the Java wrapper)
- Unified Engine loading procedure for all license types including the Online License, so that you do not need to rewrite your code when switching from one license type to another (InitializeEngine, IEngineLoader::InitializeEngine)
Please visit our website at www.abbyy.com for the most up-to-date information about ABBYY FineReader Engine and other ABBYY products.
See also
17.09.2024 15:14:41