|
|
|
|
|
|
|
Powerful OCR helps win the race for business.
ABBYY Mobile OCR Engine enables developers to integrate Optical Character Recognition (OCR) into mobile and small-footprint applications. This powerful software development kit (SDK) enables images and photographs to be transformed into searchable and editable document formats and supports all of the most popular mobile platforms and devices – including iOS and Android.
High Quality Mobile Recognition
The ABBYY Mobile OCR Engine is based on ABBYY’s world-renowned OCR technology – providing highly accurate text recognition from mobile devices. Its sophisticated functionality includes:
Automatic image skew correction. Photographs taken with a mobile device often suffer image skew, which negatively impacts recognition quality. The Engine enables detection and correction of skews within one degree of precision (the maximum detectable skew angle is 16 degrees), significantly improving the quality and accuracy of mobile OCR.
Document orientation detection function. Image pre-processing automatically detects the orientation of a page.
Hyphenation support. If the engine encounters part of a hyphenated word on one line and the second part on the next line, it will join them into one word.
Confidence level indicator. This function shows the level of certainty for recognized text, allowing developers to set flexible criteria for implementation of proofreading and verification functions.
Improved data analysis algorithm. The Engines’ image-processing algorithm enables data analysis that discards all unnecessary information in an image, boosting recognition accuracy.
Spell checking during text recognition considerably improves the quality of output.
Speed up during binarization stage. A new binarization algorithm speeds the processing time of small documents in European languages (for example, business cards) up to 10-15%. Binarization itself is seven to eight times faster.
Zonal OCR, which enables applications to recognize text blocks that have been manually set up on an image.
Low Resource Requirements
The ABBYY Mobile OCR Engine is based on compact code OCR technology and is optimized to work with devices that have small memory sizes – including smartphones, tablets and portable scanners. Features include:
Upgraded memory management. A new algorithm for memory management enables the software to determine the exact memory size required to process an image. This eliminates the need to allocate significant memory segments in advance, which can impact recognition speed and reliability – ensuring efficiency and fast performance speed.
The Engine’s code is very compact. It occupies as little as 8 MB of ROM and 10 MB of RAM depending on desired functionality.
Parallel Processing
Recognition operations are performed in parallel. The number of threads used for recognition is equal to four by default. It is possible to set up custom number of threads. Speed enhancements will be most noticeable for documents containing many lines of text.
Business Card Recognition
The Engine also processes business card images obtained via mobile device cameras or portable scanners. This allows retrieval of information including first name, last name, title, phone number, e-mail, address, etc. Plus, the technology can recognize cards in 26 languages.
Barcode Recognition
Recognition of most popular 1D and 2D barcodes is supported. See The Full List of Supported Barcodes. Several barcodes on an image can be detected.
Sixty Two Recognition Languages
Text recognition is supported for 62 languages. This includes:
23 main languages with dictionary support.
39 additional languages with Latin, Cyrillic, CJK or Greek characters.
Innovative Data Analysis Algorithm
The data analysis algorithm of the Engine enables recognition and retention of the source document’s original formatting in the output text:
Preserves multi-column text. The Engine’s Paragraph Assembly function identifies text-block borders, recognizing each block separately – preserving the format of a multi-column text, paragraphs, and text segments.
Preserves Character Fonts. The Engine identifies the font properties of source text – bold, italic or underlined.
Two Recognition Modes
One of two mobile recognition modes can be selected:
Fast mode. When an image is of good quality, this mode cuts the time required for recognition and processing.
Full mode. Best for low-quality images, when more time is required to achieve optimal results.
Step 1: Image Import and Processing
An image is loaded from memory and prepared for OCR. Image binarization separates text from the background, producing a black-and-white image that is much smaller in size than the color original. Additional skew correction and document orientation detection can be applied.
Step 2: Document Analysis
Document Analysis is a set of algorithms that then analyse the image. It detects letters, joins the letters into words and then into lines of text, and finally, into paragraphs. Additionally, the reading area is cleaned and noise removed.
Step 3: Optical Character Recognition (OCR)
Detected blocks on the image are recognized using special language and pattern definitions. If dictionaries are available, then the text is also compared to them to improve overall recognition quality. Additionally, each character is assigned a confidence level – showing how confident the recognition engine was in its final choice of character.
Step 4: Result Processing
Recognition results can be processed and exported. The developer has full control over the OCR results.
© Copyright 2000-2023 COGITO SOFTWARE CO.,LTD. All rights reserved
