US Bank Check OCR

The Python OCR SDK supports the Bank Check API.

Using the sample below, we are going to illustrate how to extract the data that we want using the OCR SDK.
Bank Check sample


from mindee import Client, PredictResponse, product

# Init a new client
mindee_client = Client(api_key="my-api-key")

# Load a file from disk
input_doc = mindee_client.source_from_path("/path/to/the/file.ext")

# Load a file from disk and parse it.
# The endpoint name must be specified since it cannot be determined from the class.
result: PredictResponse = mindee_client.parse(, input_doc)

# Print a summary of the API result

# Print the document-level summary
# print(result.document.inference.prediction)

Output (RST):

:Mindee ID: b9809586-57ae-4f84-a35d-a85b2be1f2a2
:Filename: default_sample.jpg

:Product: mindee/bank_check v1.0
:Rotation applied: Yes

:Check Issue Date: 2022-03-29
:Amount: 15332.90
:Payees: JOHN DOE
         JANE DOE
:Routing Number:
:Account Number: 7789778136
:Check Number: 0003401

Page Predictions

Page 0
:Check Position: Polygon with 21 points.
:Signature Positions: Polygon with 6 points.
:Check Issue Date: 2022-03-29
:Amount: 15332.90
:Payees: JOHN DOE
         JANE DOE
:Routing Number:
:Account Number: 7789778136
:Check Number: 0003401

Field Types

Standard Fields

These fields are generic and used in several products.


Each prediction object contains a set of fields that inherit from the generic BaseField class.
A typical BaseField object will have the following attributes:

  • value (Union[float, str]): corresponds to the field value. Can be None if no value was extracted.
  • confidence (float): the confidence score of the field prediction.
  • bounding_box ([Point, Point, Point, Point]): contains exactly 4 relative vertices (points) coordinates of a right rectangle containing the field in the document.
  • polygon (List[Point]): contains the relative vertices coordinates (Point) of a polygon containing the field in the image.
  • page_id (int): the ID of the page, is None when at document-level.
  • reconstructed (bool): indicates whether an object was reconstructed (not extracted as the API gave it).

Note: A Point simply refers to a List of two numbers ([float, float]).

Aside from the previous attributes, all basic fields have access to a custom __str__ method that can be used to print their value as a string.


The amount field AmountField only has one constraint: its value is an Optional[float].


Aside from the basic BaseField attributes, the date field DateField also implements the following:

  • date_object (Date): an accessible representation of the value as a python object. Can be None.


The position field PositionField does not implement all the basic BaseField attributes, only bounding_box, polygon and page_id. On top of these, it has access to:

  • rectangle ([Point, Point, Point, Point]): a Polygon with four points that may be oriented (even beyond canvas).
  • quadrangle ([Point, Point, Point, Point]): a free polygon made up of four points.


The text field StringField only has one constraint: its value is an Optional[str].

Page-Level Fields

Some fields are constrained to the page level, and so will not be retrievable to through the document.


The following fields are extracted for Bank Check V1:

Account Number

account_number (StringField): The check payer's account number.



amount (AmountField): The amount of the check.


Check Number

check_number (StringField): The issuer's check number.


Check Position

πŸ“„check_position (PositionField): The position of the check on the document.

for check_position_elem in result.document.check_position:

Check Issue Date

date (DateField): The date the check was issued.



payees (List[StringField]): List of the check's payees (recipients).

for payees_elem in result.document.inference.prediction.payees:

Routing Number

routing_number (StringField): The check issuer's routing number.


Signature Positions

πŸ“„signatures_positions (List[PositionField]): List of signature positions

for page in result.document.inference.pages:
    for signatures_positions_elem in page.prediction.signatures_positions):


Join our Slack