CLOVA OCR FAQ

Available in Classic and VPC

You can get your questions answered quickly by referring to the answers in the FAQ before reading the guide. If you haven't found the answer to your question in the FAQ below, then search the guide for what you would like to know.
The CLOVA OCR FAQ covers the following topics.

• Service in general
• API linkage
• Security
• Pricing

Service in general

The following covers frequently asked questions and answers for CLOVA OCR.

Q. What are the characteristics of CLOVA OCR?

A. Having equipped with a high-performance OCR recognition model optimized for major business use with the utilization of NAVER's AI technology, CLOVA OCR can quickly process jobs by accurately recognizing documents and increasing document recognition efficiency by specifying follow-up processing workflows. In addition. it features a unique set of character-area detection and recognition technologies to comprehend various types of characters. Moreover, it provides a feature to easily create a template and specify the desired area to quickly extract only the required characters.

Q. Can I automate document processing?

A. Using CLOVA OCR, documents that have been manually classified can be automatically classified based on the similarity compared to registered templates. Document classification becomes possible without user intervention, enabling design of effective workflows.

Q. What is the difference between CLOVA eKYC and CLOVA OCR?

A. CLOVA OCR is a service that extracts features in a desired format by analyzing various images uploaded by the user. On the other hand, CLOVA eKYC is a service that only extracts features of ID cards/business registration certificates through Document OCR of CLOVA OCR and checks the authenticity through a third party certification authority.

Q. Can I set CLOVA OCR to distinguish and recognize only English letters?

A. It doesn't provide a feature to distinguish and recognize only English letters.

Q. Does it provide a feature that verifies values after recognition?

A. CLOVA OCR can verify the value of recognized text, reducing repetitive verification jobs and increasing reliability. It can also be linked with other services such as Papago to provide differentiated services to users.

API linkage

The following covers frequently asked questions and answers related to API linkage.

Q. To use CLOVA OCR in the company's internal network, the IP needs to be registered in the firewall. How do I find the IP information to be registered in the whitelist?

A. You can get the IP information by configuring a VM to call. For more information on the VM proxy linkage, refer to the CLOVA OCR VM proxy linkage guide of the CLOVA OCR API guide.

Q. I want to control with the API Gateway invoke URL's IP information. How do I get the information for access control?

A. You can get the access control information by configuring a VM to call. For more information on the VM proxy linkage, refer to the CLOVA OCR VM proxy linkage guide of the CLOVA OCR API guide.

Q. The secret key needs be used to call CLOVA OCR from the client. However, the key can be exposed, causing a security problem. Is there a way to use the secret key without exposing it?

A. You can keep the secret key from being exposed by configuring a VM to call. For more information on the VM proxy linkage, refer to the CLOVA OCR VM proxy linkage guide of the CLOVA OCR API guide.

Security

The following covers frequently asked questions and answers related to security.

Q. Is there any risk that images or documents sent to CLOVA OCR could be stored or utilized separately?

A. CLOVA OCR receives the target data through the API, recognizes it from the OCR service engine, and then returns the result value text. Therefore, images or documents sent by the user are not stored in the cloud, and they are not used for the purpose of improving engine performance or analyzing patterns using data. Returned result value text is not included in the log data, and it is only stored for the OCR call access history. In other words, the service doesn't store the original recognized by OCR and recognition result, and only saves the minimal log data for billing confirmation based on call access history.

Q. Are there any security features such as encryption for text results returned through CLOVA OCR?

A. CLOVA OCR is linked through API Gateway. When using CLOVA OCR, the section between the user's server or application and API Gateway of NAVER Cloud Platform uses HTTPS communication, so the encryption of the communication section is applied. In addition, since CLOVA OCR processes original images in memory and returns the results, it doesn't store images separately. It doesn't store the recognition results either, so it can be used safely.

Pricing

The following covers frequently asked questions and answers related to pricing.

When I use the table extraction feature in CLOVA OCR, how are the fees charged?

A. If a "table" format is recognized in the uploaded image, the fees for table extraction occurs along with the text extraction fees. If no table is recognized, then only the text extraction charges will occur.