Can Abbyy OCR SDK can detect Receipts with more than 1 table?
And If it is not, Can I change it by modifying the XSD schema of the XML?
In my case,it detects the initial tabe and the remaining table items are listed in Recognised Text...I need to list all the Items in the Recognized Items Tag..
Thanks in Advance :)
Comments
1 comment
Hi,
Sorry for the delay. Currently our Receipt Capture technology can detect only one receipt on the page. You cannot influence this, because this is the issue at the level of internal technology of analysis and recognition..
However, you can try to implement receipt processing by yourself. For example, you can use the processImage method, export results in the XML export format and parse it in order to extract data you need. The XML output file contains the information about all found text blocks, their coordinates and values, therefore you will be able to find the data of your interest and represent the output in any view you wish. The basic idea is that the field value and the field name would be close to each other in the recognized text. Thus, you could parse the XML output to find the keywords like the field names (e.g. Date, Time, Total and so on), get these keywords’ coordinates and then find other text blocks (the field values) situated somewhere near these keywords (right below them, on the same level to the right, etc.). You can use the baseline coordinate of the words for this: if the baseline coordinates are close, the words are from the same line.
Please sign in to leave a comment.