Nanonets is an AI-based workflow automation software centred around our revolutionary OCR engine. We have many pre-trained models to extract data from standard customer documents, Receipts being one of them.
We can help extract structured information from Receipts in the form of flat fields or key-value pairs, such as Total Amount, Merchant_Name, Tax_Amount, etc. as well as tabular fields.
Alternatively, in case you would want a different set of fields to be extracted, you can use our zero-training OCR model which can be set up to extract custom fields within minutes. Here is more information on how you can use the Zero-Training Model.
The list of fields that we support out-of-the box are as follows:
| Field | Description |
|---|---|
| Total_Amount | Extracts the Total Amount due to the merchant. |
| Tax_Amount | Extracts the Tax amount due to the merchant. |
| Cash_Tender | Extracts the amount paid in cash, if present on the receipt. |
| Merchant_Phone | Extracts the contact number of the merchant. |
| Merchant_Address | Extracts the address of the merchant. |
| Receipt_Number | Unique identifier for the receipt. |
| Merchant_Name | Extracts the name of the merchant. |
| Card_Tender | Extracts the amount paid with a card (debit or credit), if present on the receipt. |
| Subtax | Extracts the subtax amount, if present on the receipt. |
| Date | Extracts the date on which the receipt is generated |
You can always add more fields if you need or refine/retrain the model for better extraction accuracy on your own Receipts. We also keep adding more and more fields every month based on customer requests.