Hi, I've been using the Cloud OCR for several months with mixed results. One of the issue we have is that we have a well know shape that we need to 'scan' for and recognize the text within that specific shape. We don't have many problems getting the text using the OCR Clound but it does not always fall within the specific shape we are looking for (so we get many false results). Is there any way for the OCR cloud instance to 'learn' a shape (and scan only within the bounds of that shape)? If not are there other ABBYY products that allow us to upload an image file and have the engine learn/scan for text only within that shape? Thanks.
asked 20 Oct '15, 02:18
If we speak about ABBYY Cloud OCR SDK, you can try the following scenario: at first you process your image with the processImage method and export it to the XML export format with the image size and the text coordinates in pixels, then extract the necessary coordinates and call the processFields method. In this case re-recognition will be performed for free.
Alternatively, have a look at ABBYY FlexiCapture Engine, it is a non-cloud based Data Capture SDK designed to process the structured documents. If you are interested you can contact with your region sales for getting a trial version or simply fill the following form: https://www.abbyy.com/data-capture-sdk/request/?purpose=trial
answered 20 Oct '15, 12:11
Oksana Serdyuk ♦♦