Detectors

  • Motion Blur and Defocus Detector
  • Glare Detector
  • Noise detector

Presets

  • Black-and-white document
  • Business card
  • Color document
  • Light-on-dark document
  • Receipt
  • Spy shot
  • Stamped paper
  • Street shot

Visual Quality Enhancement

  • Manual brightness/contrast adjustment
  • Auto brightness/contrast
  • Auto enhancement
  • Auto cropping / Edge detection (Recognize edges)
  • Binarization
  • Clean background
  • Gray scaling
  • Manual crop
  • Noise removal
  • Heavy cleanup
  • Inversion
  • Rotation
  • Fixed rotation
  • Sharpening
  • Perspective correction
  • Whitepaper

Image Distortion Detectors

For customers that require the highest accuracy of text recognition on documents captured with mobile devices the SDK provides intelligent image analysis: it detects blurred photos and images taken with uneven or poor lighting that should be taken again before they are sent to an OCR engine.

Moniton Blur and Defocus Detector

Back to top

Before looking for defects, the detector divides the entire image into blocks and analyses each block to determine whether it contains any text or not. All subsequent information relates only to the blocks with text.

As a result of the analysis, the detector provides the following information:

Mobile Imaging SDK sreenshot Mobile Imaging SDK sreenshot

Before

After

Glare Detector

Back to top

The detector provides the following information:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Noise Detector

Back to top

The detector provides the following data:

You can use one of those parameters to assess if there is sufficient light on the image. The signal to noise ratio is intended for customization, while the binary value is set during the analysis.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

How to interpret the detector output

To ensure high quality text recognition, the glare and noise detectors should produce the "false" result and the defocus detector should deliver close to "zero" value for the size of the blurred and defocused areas. It is assumed that for typical images the threshold size of blurred and defocused areas should be predefined by a user as it may differ for different types of images.

Presets

Presets combine image processing functions for the most typical scenarios such as processing of color documents, black-and-white documents, business cards, receipts, stamped papers, street shots and more.

Black-and-white document

Back to top

Combines a group of functions for processing black-and-white documents. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Business card

Back to top

Combines a group of functions for processing business cards. The operations are applied in the following order:

Color document

Back to top

Combines a group of functions for processing color documents. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Light-on-dark document

Back to top

Combines a group of functions for processing light-on-dark documents. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Receipt

Back to top

Combines a group of functions for processing receipts. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Spy shot

Back to top

Combines a group of functions for processing blurry, out-of-focus, and dark pictures. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Stamped paper

Back to top

Combines a group of functions for processing stamped papers. The operations are applied in the following order:

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Street shot

Back to top

Combines a group of functions for processing street view pictures with high level of noise. Contrast and brightness are adjusted automatically.

Mobile Imaging SDK sreenshot Mobile Imaging SDK sreenshot

Before

After

Visual Quality Enhancement

Manual brightness/contrast adjustment

Image contrast can be adjusted based on the adjustment level.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Auto brightness/contrast

Back to top

Automatic tuning of brightness and contrast.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Auto Enhacement

Back to top

Enhances local contrast and converts background to white (the output image is in gray scale).

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Auto cropping/Edge Detection (Recognize edges)

Back to top

Determines the edges of a document relative to the background, so that the background could be cropped.
Provides coordinates of the area that should be cropped (four corners of the quadrangle).

 
 

Binarization

Back to top

Converts image from color/grayscale to black and white.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Clean Background

Back to top

Automatically removes background elements such as texture, dots, lines, etc.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Gray Scaling

Back to top

Converts color RGB images into grayscale.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Manual Crop

Back to top

Allows users to crop an image to a specified rectangle.
There is no special interface to manually change the image corners' coordinates in the sample application, i.e. they can be changed only through coding.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

When special interface to change the coordinates of image corners is applied to the end-user app, Manual Crop works as shown on the screenshot (from ABBYY FineScanner).

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

FineScanner Manual Crop Before

FineScanner Manual Crop After

Noise Removal

Back to top

Removes 'stray' pixels from the image based on the threshold adjustment level.

undefined

undefined

Before

After

Heavy cleanup

Back to top

Removes noise from a binary image.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Inversion

Back to top

Inverts black and white images.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Fixed Rotation

Back to top

Reorients an image through multiples of 90 degrees.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Rotation

Back to top

Rotates an image to a random degree.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Sharpening

Back to top

Removes blurriness around characters based on the adjustment level.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Perspective Correction

Back to top

Corrects perspective distortions (from trapezoidal shape to rectangular).
There is no special interface to manually change the image distortion in the sample application, i.e. it can be changed only through coding.

 
 

When special interface to change the distortion of image is applied to the end-user app, Perspective Correciton works as shown on the screenshot (from ABBYY FineScanner).

undefined

Mobile Imaging SDK sreenshot

FineScanner Perspective Correction Before

FineScanner Perspective Correction After

Whitepaper

Back to top

Makes the document’s background white.

Mobile Imaging SDK sreenshot

Mobile Imaging SDK sreenshot

Before

After

Want to try?

Contact sales team for more information or request a free trial

Usage of Cookies. In order to optimize the website functionality and improve your online experience ABBYY uses cookies. You agree to the usage of cookies when you continue using this site. Further details can be found in our Cookie Policy.