Check OCR

Automatically extract key information from bank checks using our OCR API

Get early access

What is a bank check OCR?

A check OCR stands for Check Optical Character Recognition. OCR is the technology that transforms an image or a PDF from a bank check into machine-readable data. Reading a check thanks to OCR will enable the extraction of key information such as amount, payees, date etc... The data extracted can then be sent automatically to specific software thanks to APIs or libraries. Thanks to this process, users will be free from manually entering the information written on the check. This help reduce both errors and friction.

OCR - Optical Character Recognition

OCR is the name of technology that enables the detection and the reading of a text in an image. It will transform the text from the image into machine-readable data. If you are curious to learn more about OCR, take a look at this exhaustive blog post.

Bank check OCR vs OCR

When used alone and in a generic context, OCR refers to the action of extracting and reading words in an image. When referring to a specific type of document, for instance Invoice OCR or Bank check OCR, this will involve extracting specific information out of the document. This process is called key information extraction. For instance, Bank check OCR is referring to the process of extracting the amount, the payees information and more.

Bank check scanning

Checks can be both digital and paper. For paper check processing, the first step of the process is the scanning. That involves taking a picture of it so that it can be sent to a specific OCR for key information extraction.

MICR line

MICR stands for Magnetic Ink Character Recognition. This technology is used in the banking industry on check for processes and clearance purposes. On bank checks, the MICR line contains the account number, routing number and check number information. The check OCR API includes a specialize MICR OCR able to read those special characters.

Check OCR extracted fields

Transform any bank check - paper or digital - into usable data in your software. Free your users from manual entries, remove friction and reduce errors.


The amount written on the check in the currency indicated. The amount can be typed or handwritten. Both are supported.


The list of person names to whom the check is made payable. If multiple payees are indicated then each payees‘ name is extracted.

Check issuance date

The date at which the check has been issued. 
The date is returned in ISO format YYYY-MM-DD.

MICR line

The MICR lines contains the following codes encoded wit the special MICR characters:

- Routing number
- Account number
- Check number

Check’s position

The coordinates of the check’s position within the image.

Signatures’ position

The coordinates of the check signatures within the image. Multiple signatures are supported.

Check OCR - FAQ

Assessing OCR solutions requires an analysis of a couple of critical criteria such as performances for both extraction and response time, ease of integration, pricing... To help you in this process, we have gathered answers to the questions you might have during evaluation. Please reach out to us using our chat if you have further questions regarding our check OCR API.

📄 How can I test the bank check OCR API?

The bank check OCR API is in private beta. If you are interested in testing it, request an early access.

A member of our team will organize a demo and answer any questions you might have.

💸 What is the pricing?

Start trying the check OCR API thanks to the free plan that includes 250 pages monthly.

For monthly consumption above 250 pages, the pricing starts at $0.10 per page and is decreasing gradually. Check our pricing page to see available plans or reach out to our team to know more.

🗺️ What are the supported countries?

Our technology is based on Computer Vision, which means that it relies on image analysis rather than text analysis to extract bank check data. As such, there is no language limitations to the check OCR API.

🖋️ Does it support handwritten fields?

Handwritten fields such as amounts are extracted.

For signatures, the position of the signature on the check is detected and outputted in the API response. If there are multiple signatures on the document, the different positions will be included in the response.

📈️ Does the API support MICR characters?

The MICR Line uses special character to indicate routing number, account number and check number on the check. This 3 MICR codes are used for processes and clearance purposes.

Our models have been trained to support those characters thanks to a specialized MICR OCR. Those 3 fields are available in the API response.

🎯️ Does the OCR work on low-quality image?

Bank check OCR works on digital bank checks and also paper bank checks that are scanned. Our models are trained on a wide variety of documents to make our API robust to various image quality.

To take into account potential ink stains on the check or blur on the scan, we train our models including augmentations. The OCR will be able to read most of the documents as long as they are readable.

⚡ How difficult is it to integrate the API?

The integration is seamless and straight forward. There are different ways of integrating and accessing our API products.

We recommend using our client libraries available in Python, Node, Ruby and Java.

A custom integration can be made using our REST API.

🤓️ Does Mindee offer technical support?

If you have any questions, feel free to reach out to our team as we offer a wide range of support:
- Access our API and client libraries documentation
- For technical questions regarding integration get help from our slack community
- For pricing plan and commercial questions use our chat and meet with our team

Check OCR - use cases

Checks are involved in numerous use cases as it is a reliable and secure way of payment, whether in paper or digital. The MICR technology helps with clearance process, making it a preferred choice for certain type of transactions such as rent payment or claim refund. Let’s take a look at a couple of examples in various industries.

Insurance - Claim management

In the Insurance Industry, checks are issued and processed for claim management purposes. Claim Management softwares need to process checks coming from Insurance companies to settle claims. The check will be read to verify that its amount matches the claim settlement. The payment can then be made using the list of payees on the check. Thanks to the check OCR API, those softwares can automatically access those key information from the check and reduce processing time.

Financial services - Check cashing

For the banking industry, the main application is check cashing. For paper check, the first step of the deposit process is to take a picture of the check. The MICR line reading will trigger the clearance process. Then the reading of the amount and the payees will help initiating the transaction. The money can be moved from the payer’s account to the payees’s account. The check OCR API that includes a specialized MICR OCR facilitates the overall process.

Try out our Bank Check OCR API

Try out our products for free. No commitment or credit card required. If you want a custom plan or have questions, we’d be happy to chat.

logo Mindee

Schedule a meeting with one of our experts

Please provide the following information so we can connect you to the right teammate.

Oops! Something went wrong while submitting the form.