OCR Reader

Available in Classic and VPC

For nondevelopers to be able to use them easily, OCR Reader provides General, Template domains in a web UI form that are created through CLOVA OCR. To use OCR Reader, you need a storage. Storage can be created using Object Storage from NAVER Cloud Platform. For more information on how to create a bucket with Object Storage, see Object Storage user guide.

OCR Reader interface

The basic description for using OCR Reader is as follows:

Check folder list

You can check the information on each folder from the OCR Reader list. To view:

1. In the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA OCR.
2. Click the OCR Reader menu.
3. When the folder list appears, check the basic information.
   • Name: Folder name
   • Recognition type: Recognition settings information
   • Size: Size of all documents in the folder
   • Date of creation: Date of folder creation

Folder creation

To create a folder:

1. In the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA OCR.
2. Click the OCR Reader menu.
3. Click [Create recognition folder].
4. Enter the required information.
   • Recognition type: Choose the type of document recognition.
     • Text/table: Extract all texts and tables that the image includes.
     • Template: Extract texts or values that the customer recognizes in the designated area.
   • Create a recognition folder: Detailed settings of a recognition folder
     • If texts/tables are selected
       • Folder name: Enter only lowercase English letters and numbers up to 25 characters.
       • Folder location: Information on Object storage bucket where you recognize and save images
         • Automatic creation: Create a bucket randomly in Object Storage.
         • Manual selection: Designate a bucket directly in Object Storage.
       • Supported languages: Select among Korean, Japanese, and Chinese.
       • Table extraction: Decide whether to extract tables. To extract tables, you have to activate table extraction in the selected domain.
     • If you selected a template
       • Folder name: Enter only lowercase English letters and numbers up to 25 characters.
       • Folder location: Bucket information of the Object Storage where you recognize and save images. Even if you delete the folder, the folder location remains.
         • Automatic creation: Create a bucket randomly in Object Storage.
         • Manual selection: Designate a bucket directly in Object Storage.
       • Domain ID: Setting of the domain where you implement recognition (domain ID/domain name/domain code)
       • Automatic classification of templates: Among templates in the domain that you designated in the domain ID, choose one you want to recognize.
5. Click [Create].

Upload recognition files

To upload recognition files:

1. In the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA OCR.
2. Click the OCR Reader menu.
3. When the folder list appears, doubleclick the folder you want to upload.
4. Click [Upload recognition file].
5. Drag the file you want to upload or select an image from the upload popup.
6. Click [Apply].
7. Click [Implement recognition].
   
8. Check the recognition results.
   

Edit folders

To edit folder information:

1. In the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA OCR.
2. Click the OCR Reader menu.
3. Click the folder you want to edit, and then click [Set folder].
4. Edit the necessary information.
5. Click [Apply].

Delete folders

To delete a folder:

1. In the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA OCR.
2. Click the OCR Reader menu.
3. Click the folder you want to delete, and then click [Delete].
4. When the Delete popup appears, click [OK].