Recognize Date and other similar fields from image using Aspose.OMR for .NET

Hello!!
how to handle Date ?
Can we do something like this (check attached image)? in OMR.NET?
Can we control density of checkboxes ?
image.jpg (256.6 KB)

@onlyformydev

Your inquiry has been logged under the same ticket OMRNET-78 in our issue tracking system. We will let you know as soon as some updates are available. Please spare us some time.

@onlyformydev

  1. Date and other similar fields (ID, for example) can be handled by Grid element, that covers two dimensional input fields and provides single result.
  2. Not sure exactly what you would like to do: process this particular template or generate new one with same structure. So we cover both cases:

a) We’ve created template for your image using GUI tool Aspose.OMR Client, printed it, took a couple of photos and recognized them with Aspose.OMR for .NET as a demonstration of Aspose OMR capabilities.

Please check attached archive, it contains two folders:

  • template, which contains template files: .omr file with questions markup and empty image that is used for printing new forms
  • photos, with two photos of partially filled forms and recognition results in .csv format b) In case you would like to create similar template using Aspose.OMR generation feature - we provide Text, Grid and ChoiceBox elements, and you can customize almost everything in these elements. For example, you can customize the amount of options in ChoiceBoxes, the type of answers (ABCD,1234,etc.) and their position relative to other elements. So, while you can’t create the same form, you can create structurally very similar template using our generation feature with same questions and text.
  1. While you can’t directly control density of choice boxes, you can use Sheet to generate columns and rows of questions that are located close to each other. Please check our docs for examples of images and markups: Designing machine-readable forms|Documentation

Aspose.zip (6.3 MB)

Hello!!

[quote=“asad.ali, post:3, topic:210211”]
Not sure exactly what you would like to do: process this particular template or generate new one with same structure. So we cover both cases:
[/quote]

Both

Where I could Find this product … ?

Does Template creating is Still Text based ?

So template Generation from image use OMR cloud … ?

Can U please explain how whole process works ?

There are multiple thoughts in my mind.

1
#DD
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)
#MM
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)
#YY
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)
(0) (1) (2) (3) (4) (5) (6) (7) (8) (9)

but when <*> is added in front it stop working.
at least 2 options needed to be selected.

2
if possible parentheses should include more then on character/s in it like

#DD
(1) (2) (3) (4) (5) (6) (7) (8) (9)
(10) (11) (12) (13) (14) (15) (16) (17) (18) (19)

3

?grid=Day
sections_count=3

?grid=Month
sections_count=2
?grid=Year
sections_count=2
It fill whole page vertically
But No Way to Align these grids in single row.

@onlyformydev

It is possible to process any templates using Aspose OMR. To do so, we created free WPF tool - Aspose.OMR Client, that can be used to create templates (.omr files) from any images. It is located at our GitHub: https://github.com/aspose-omr-cloud/aspose-omr-cloud-dotnet/tree/master/Aspose.OMR.Client. After downloading the Client and loading chosen image, user should create mark-up by placing visual elements (such as ChoiceBox and Grid) over questions on the image and setting properties (answer options, question name, etc.).

We have logged these details along with the ticket and will surely inform you as soon as we investigate further. Please spare us some time.

@onlyformydev

  1. Indeed it can be done using simple questions. In current version recognition provides all marks that were recognized, not taking into account multi-select properties. So feel free to use # to start a question. In future version with grading support included, we will support multi-selection properties properly.
  2. Currently only single characters are supported for bubbles. However, we will support two-chars bubbles in the future.
  3. Indeed, for the moment grid can only be aligned vertically. Horizontal grid alignment is also planned in the future.

So, for your needs you could use choice-boxes (starting with #). The only inconvenience that may be caused is that in choice-boxes all recognized answers are separated using special characters, so you may need to process that additionally. However, we will provide support for mentioned features in future releases.