OMR

OMR (Optical Mark Recognition) allows checkboxes and marks recognition of any form and shape by using preconfigured zones on the target marks.

The recognition is done by creating (drawing) a zone on the target mark, including or not the outer borders depending on the zone settings. The recognition will detect if that mark is filled or not and return its status as result. This process is very useful also to detect if any type of boxes are filled with something, for example signatures.

By default the module creates automatically variables for every zone inserted with the naming convention, for every zone, of:

  • OMR_ZONEXX

Where XX is the incremental number of the zone starting from 0. The variable name can be further customized inside the zone setting.

When processing multipage documents every variable will be extracted on every page and the variable name will be suffixed with _PY which stands for Page Y where Y is the number of the page, example:

OMR_ZONE01_P1
OMR_ZONE02_P2

The variable name without the _PY will still exists and containing the result on the last page of the document.

The module creates custom variables during processing. Check the Variables list for more details.

The left hand side menu shows the available settings section. Settings are displayed according the selected section.

General

processing_settings_omr1

This section shows a sample built-in document preview, moreover it allows to load real document samples in order to draw and create the target recognition zones.

On the top bar document controls are available:

Info

Image controls apply to the preview only and not to the physical document stored in the repository.

Load Image
Browse to open and load (by uploading to the server) a sample document to render on the preview pane.

Info

Browse allows to upload the sample document to the server, this operation stores the document file in the local Scanshare data folder.

The sample document will be saved inside the module settings and reopened the next time the module is configured by using the document file uploaded to the Scanshare data folder.

If the document file has been removed or the Workflow has been exported on a new machine where the sample document file is not existing the preview will render with a big red cross X, however zones are existing and fully working unless they are modified.

Perform test recognition on inserted zones
When pressing this button, the application will start recognition all configured zones. When all Zones are read, a notification message will be prompted and the zones values will be updated in the zones list in order to check the recognition test results.

Draw recognize area
With this option, you can manually select an area where the application should read the Zone from. When a Zone is detected, it will automatically be read and show up in the zones menu with the recognition result.

Draw anchor
With this option, you can draw a zone anchor on the document which is colored green rather than yellow as the standard zones.

The Anchor is a special function to verify and correct shifting delta when scanning the documents in order to adjust the zone coordinates considering the scanner shifting.

Anchor can be anything distinguish on the document such as: a customer logo, a header row, a footer section and so on. Make sure the select something which appears on every scanned document and which is good comparable and not mistaken with any other components appearing on the document. It is suggested to don’t include much white space around the real contents on which the zone is drawing to.

The process saves the Anchor zone on the sample document used assumed to be the reference document. For every scanned document the module will search the contents of the Anchor zone and, once located, compare the position to the reference image zone position. Any shifting found between the reference position will be counter applied to the zones position too.

Info

When using the Anchor make sure the load a production good scanned document in order to have an accurate anchor zone position reference.

Delete the selected object
Delete the current selected zone.

Zones can be deleted also with the DEL keyboard key, once selected.

Fit
Fit the document to the current view.

One to One
Enlarge the document to the 100% zoom ratio into the current view.

Zoom In
Zoom In the document into the current view.

Zoom Out
Zoom out the document into the current view.

Rotate Clockwise
Rotate clockwise direction the document by a manner of 45 degrees.

• Rotate Counter Clockwise
Rotate counter clockwise direction the document by a manner of the 45 degrees.

• Page
Displays the current preview page number on the total number of pages of the document.

• Previous page / Next page
Load and display the previous or next page preview of the document.

Abbyy toolbar

On the bottom bar Abbyy engine controls are available:

processing_settings_omr3

• Group distance
Select a drawing tool to manually draw an area on the distance between two group zones, e.g. between the first answers group and the second one. The area will be highlighted in red.

• Row distance
Select a drawing tool to manually draw an area on the distance between two marks of the same group, e.g. between the first answer and the next one, horizontally or vertically. The area will be highlighted in dark red.

• Training empty mark
Select a drawing tool to manually draw an area on a custom mark in order to teach to the engine its custom type.

Info

For best approach the distance space needs to take care of white space bounds as well of the group itself. It needs to start after the group zone finishes and ending considering this space as well in the next row.

Engine

This section allows to customize general engine settings and recognition behavior for the current module.

Engine
Select here the OCR engine to use to run the current recognition. Available engines are, based on the current license:

  • Nuance OmniPage
  • Abbyy FineReader

Based on the selected engine further settings will be displayed.

OmniPage

processing_settings_omr_nuance

Type
Select the type of zone recognition to apply. Available types are:

  • Auto: automatic detection of the markup frames (borders) inside the zone
  • WithFrame: the zone contains the frames (borders) – means the zone has been token around the complete box
  • WithoutFrame: the zone doesn’t contain the frames (borders) – means the zone has been token inside the markup box

Sensitivity
Select the level of sensitivity of the zone recognition. Available types are:

  • Highest
  • High
  • Low
  • Lowest

Abbyy

processing_settings_omr_abbyy

Zone Type
Select the type of zone which applies to the current area. Available options are:

  • Single: standard single OMR zone. It is drawn around the single checkbox.
  • Group: group OMR zone. It is drawn around a single “question” including all checkboxes which are part of the answers, both vertically or horizontally.

The Single OMR zone is the standard zone which is returning the status of the mark, either filled or not. The Group OMR zone is a special zone including all answers (marks) for a certain question, it is returning only the filled marks and if a Mask is provided it will return just the exact answer(s) corresponding the filled mark(s) based on the provided mark.

Checkboxes count (only for Group)
Number of checkboxes which the Group zone is including.

Direction (only for Group)
Direction of the answers / zone. Available options are:

  • Horizontal: the answers are on the same row and the zone expands horizontally.
  • Vertical: the answers are on the same column and the zone expands vertically.

Maximum checked (only for Group)
Maximum number of checkboxes which can be found marked.

Minimum checked (only for Group)
Minimum number of checkboxes which can be found marked.

Rows count (only for Group)
Total number of rows existing matching the drawn group zone (e.g. number of questions one below another one).

Mask (only for Group)
Mask of the results to use during result checking, with values separated by (,) comma.

Assuming as an example the question below with 5 answers named from 1 to 5:

 processing_settings_omr5

Providing a mask of 1,2,3,4,5 the result returned in the variable after the recognition will be 1, the filled mark corresponding the provided mask.

Info

Make sure the mask has the same number of values of the checkboxes inside the group otherwise an error will be generated.

Checkbox type (only for Single)
Select the type of the checkbox for this zone in order to have the best recognition accuracy. Available options are:

  • Empty
  • Circle
  • Square
  • Custom

By selecting Custom checkbox type a training of the current mark shape is required by using the bottom bar toolbar button Training empty mark.

Enhance local contrast
If enabled engine will increase the local contrast of the image during the preprocessing of the image. Such option may increase the quality of recognition.

Info

The option is meaningful for color and gray images only.

The images for which this preprocessing method is effective include:

  • Photos or scans of documents with texture or pictures in the background. With the normal binarization procedure, the characters that coincide with darker areas of background may be lost or recognized unreliably. If you apply this method before recognition, such areas are detected, and contrast is increased, with the result that after binarization the characters stand out more distinctly.
  • Photos or scans of documents with highly colorful background or text highlighting.

Remove noise
If enabled engine will reduce the noise of the image. Available working options are:

  • White noise: this mode may be useful, for example, for uncompressed images with ISO less then 800, for reduced images.
  • Correlated noise: this mode may be useful, for example, for the JPEG photos with high compression settings

Zones

The table shows a list of all configured zones.

The view of the table is very easy and intuitive:

• Variable
The variable assigned to the zone which will contain the output result of the recognition.

• Text
The recognition result populated when testing or adding the zones.

• Id
The id associated with your zone. It is unique and it is the id of your zone inside the DB.

• Status
If the zone is enabled or not, a disabled zone will not be used. The status of a new zone is automatically set to true when a new zone is created. In the zones menu you can easily set this to false by clicking on the indicator. The zone will then be greyed out, the indicator will be gray and set to off and it will be skipped during the recognition.

Zone

processing_settings_omr2

Variable
Enter the name of the variable which will contain the recognition result of this zone. If the variable is already existing in the current Workflow an error message will prevent the zone to be saved.

Type
Select the type of zone recognition to apply. Available types are:

  • Auto: automatic detection of the markup frames (borders) inside the zone
  • WithFrame: the zone contains the frames (borders) – means the zone has been token around the complete box
  • WithoutFrame: the zone doesn’t contain the frames (borders) – means the zone has been token inside the markup box

Sensitivity
Select the level of sensitivity of the zone recognition. Available types are:

  • Highest
  • High
  • Low
  • Lowest

All pages
If selected the recognition will run automatically on all pages. When this option is enabled the current variable will always contain the last page result while an automatic variable for every page is created in the form of:

VARIABLENAME_PX

Where the current variable name, either automatic generated or customized, will be suffixed by _PX where X is the number of the page.

Only on page
Enter the number of the page where the current zone needs to be extracted. This automatically contains the number of the page of the sample document where the zone has been drawn.

Last page only
If selected the recognition will run only on the last page of the document, whatever number of pages the current document has got.

Previous Article

MICR

Next Article

Report Generator