OCR Engine 2: Issue with Danish characters

OCR.space Free OCR API
1

I am using OCR Engine 2 to scan quotes from a book in Danish (Engine 1 did not provide meaningful output). The text is generally picked up nicely, but unfortunately the Danish characters æ, ø, å are generally not handled correctly. For example, the character å is interpreted as ả; the character æ is interpreted as either ae or a; and the character ø is interpreted as o.

Do you have plans to implement support for the Danish language for OCR Engine 2? (I can select Danish for Engine 1, but not for Engine 2). Also, any workarounds be appreciated.

I am using an iPhone 13 Pro to take the photos, and this provides me with sharp images in a 3024 × 4032 resolution.

1 Like
2

Hello, unfortunately you are correct. OCR Engine1 supports Danish, but the OCR quality is often not as good as OCR Engine2.

We just added six new OCR languages to Engine2. We plan to add support for Danish in the future, too.

1 Like
3

Thanks, I am happy to hear that you will add support for Danish at some point. In the meantime I guess I will create a text macro to correct the most common mistakes.

4

Danish OCR is availabe with OCR Engine3 now :slightly_smiling_face:

For example, the OCR result for this image:

danish ocr test image

is this text:

meaningful output). The text is generally picked up nicely, but unfortunately the Danish characters æ, ø, å are generally not handled correctly. For example, the character å is interpreted as å; the character æ is interpreted as either ae or a; and the character ø is interpreted as o.