Does the OCR Model Support Tables/Line Items

Yes our OCR Model supports Line Items and Tables.  Below is documentation on how Line Items and Table Extraction works.

What is Table and Line Item Extraction and how does it work?

Nanonets is able to identify Tabular structure in a File and extract the entire table as well as assign each column the labels you are looking for. In case a column doesn't belong to any label of interest it isn't assigned any label.


Video Demo:




Is it possible to train it learn my columns of interest?

Yes. You can train the model to identify the columns of interest and map it to the correct label as well

You can specify each of column label mappings and the model will learn to identify which column belongs to which label and make accurate predictions. Below is a video of the entire process.


Video Demo:



What does the JSON response look like for Tables?

[
   {
      "label":"",
      "score":99.94733,
      "xmin":150,
      "xmax":2401,
      "ymin":1299,
      "ymax":2505,
      "type":"table",
      "cells":[
         {
            "row":1,
            "col":1,
            "row_span":1,
            "col_span":1,
            "label":"Quantity",
            "xmin":150,
            "ymin":1299,
            "xmax":323,
            "ymax":1362,
            "score":0.9994733,
            "text":"12 ",
            "status":""
         }
      ]
   }
]


What is the Pricing for Line Item Extraction?

You can see more details here: How is the Price of Lines Items Calculated

Was this article helpful?