Recognition

Aspose.OCR for Java can extract text from a wide variety of file formats and media sources.

Extracting text from images

• Extracting text from an image
  Reading text from raster images in JPEG, PNG, WBMP, BMP, and GIF formats.
• Extracting text from multi-page TIFF
  Reading text from multi-page TIFF images.
• Extracting text from pixel array
  Reading text from images, provided as an array of pixels.
• Fast recognition
  Reading images in fastest recognition mode that consumes minimum possible resources.
• Recognizing single line
  Reading text from images containing a single line of text.
• Extracting text from receipts
  Digitizing scanned receipts without manual retyping.

Extracting text from documents

• Extracting text from PDF document
  Reading text from a PDF document that consists of scanned images without searchable text.

Extracting text from alternative media

• Batch recognition
  Reading text from a list of raster images, folder, or ZIP archive.
• Extracting text from URL
  Reading text from raster images hosted on web sites.

Identifying recognition problems

Non-fatal recognition errors are stored as a list of strings in the warnings property of the recognition result.

AsposeOCR api = new AsposeOCR();
RecognitionSettings recognitionSettings = new RecognitionSettings();
recognitionSettings.setLanguage(Language.Ukr);
RecognitionResult result = api.RecognizePage("source.png", recognitionSettings);
// Show recognition errors
result.warnings.forEach((w) -> {
	System.out.println(w);
});