Nicomsoft OCR SDK – List of Features
- Cross-platform OCR: Windows and Linux platforms are supported
- Supports 26 languages: Bulgarian, Catalan, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Hungarian, Indonesian, Italian, Latvian, Lithuanian, Norwegian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, Turkish
- Supports many image formats: BMP, JPEG, PNG, TIFF, GIF and many others
- Supports multipage image formats (TIFF, GIF)
- Supports PDF files (by default GhostScript is used)
- Ability to save recognized document as PDF (built-in PDF generator), RTF or TXT file
- Ability to load image from a file, memory or raw pixel data
- Ability to scan documents: both TWAIN and WIA interfaces are supported (Windows version only, not supported in Linux version of OCR SDK)
- Bar codes support: EAN-13/UPC-A, UPC-E, EAN-8, Code 128, Code 39, Interleaved 2 of 5 and QR codes.
- Advanced deskew algorithm
- Ability to detect misoriented pages (90/180/270 degrees rotation) and fix it automatically
- Ability to invert/rotate/mirror entire image or some text block(s) before processing
- Ability to scale image automatically for better recognition
- Robust adaptive image binarization supports images with poor brightness or low contrast
- Ability to configure image binarization parameters for specific images
- Advanced page layout analysis algorithm (zoning)
- Ability to perform OCR step by step and get intermediate results
- Advanced lines detection and removing algorithm.
- Ability to use several CPU cores for even faster image binarization
- Ability to perform zonal OCR: select one or multiple areas for OCR
- Ability to specify different OCR options for different areas
- ICR module for handwritten digits
- Ability to specify multiple languages for OCR
- Ability to specify different languages for different areas on an image
- Ability to detect and handle inverted text
- Ability to use several CPU cores for even faster OCR, even if OCR is performed for one page only
- Thread-safe: Ability to process multiple images at once by using multiple threads.
- Unique character analysis technology delivering reliable recognition of any fonts
- Advanced algorithms for poor-quality text, distorted, connected and broken characters
- Dictionaries are used for the best recognition
- Ability to use user-defined dictionaries
- Ability to format text automatically: remove unnecessary line breaks, combine divided words, detect marked lists, etc
- Many optimizations with SSE CPU instructions for fast OCR
- Ability to change default options and improve OCR results if some parameters of image are known
- Ability to exclude some characters from character set when necessary
- Ability to get additional information about text lines, words and characters: position, size, quality, etc
- Ability to specify regular expressions to improve recognition of formatted data
- Full Unicode support
- Mobility: OCR binaries and data files are less than 10MB for one language.
- Simple API interface and sample projects for various programming languages: C#, C/C++, VB.NET, Delphi, C++ Builder, Visual Basic
- A wide set of supported frameworks and technologies that can use Nicomsoft OCR SDK: .NET, WPF, WCF, ASP.NET, Silverlight, etc
- Both native x86 and x64 binaries are available (Windows version only, in Linux version of OCR SDK only x86 binaries are available)
Download a trial version of Nicomsoft OCR SDK.