Optical Character Recognition

Zia Optical Character Recognition electronically detects textual characters in images or digital documents, and converts them into machine-encoded text. Zia OCR can recognize text in 10 major languages.

Note: Catalyst does not store any of the files you upload in its systems. The files you upload are used for one-time processing only. They are not used for ML model training purposes either. Catalyst components are fully compliant with all applicable data protection and privacy laws.

OCR

Description

This API is used to detect textual characters in images and documents, and deliver the recognized text as a JSON response. The response also contains a confidence score, which defines the accuracy of the detection.

You must specify the path to the image or document file in the API request, as shown in the sample request. You can optionally specify the languages present in the text, for quicker processing. OCR supports 9 international languages and 10 Indian languages, that are mentioned in the tables below. The language is automatically detected and the text is processed, if it is not specified.

Request URL

https://api.catalyst.zoho.com/baas/v1/project/{project_id}/ml/ocr

project_id - The unique ID of the project

Request Headers

Authorization: Zoho-oauthtoken 1000.910***************************16.2f****************************57
content-type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW

Request Method

POST

Scope

scope=ZohoCatalyst.mlkit.READ

Form-Data Properties

Parameter NameData TypeMandatoryDescription
imageFileYesThe input file to be processed

You must provide the path to it in your local system

Allowed formats: .jpg, .jpeg, .png, .bmp, .tiff, .pdf
File size limit: 20 MB
languageStringNoThe language code of the text to be identified

Refer the tables below for the language codes

International Languages Supported by OCR

LanguageLanguage Codes
Arabicara
Chinesechi_sim
Frenchfra
Italianita
Japanesejpn
Portuguesepor
Romanianron
Spanishspa

Indian Languages Supported by OCR

LanguageLanguage Codes
Englisheng
Hindihin
Bengaliben
Marathimar
Telugutel
Tamiltam
Gujaratiguj
Urduurd
Kannadakan
Malayalammal
Sanskritsan

SDK documentation

OCR- Java SDK

OCR- Node.js SDK

Sample Request: OCR


				curl -X POST \
  https://api.catalyst.zoho.com/baas/v1/project/4000000006007/ml/ocr \
  -H "Authorization: Zoho-oauthtoken 1000.910***************************16.2f****************************57" \
  -H 'content-type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW' \
  -F 'image=/Desktop/HelplineCard.jpg'  \
  -F 'language=eng,spa'			

Sample Response: OCR


				{
"status":"success",
"data":{
"confidence":79.71514892578125,
"text":"Whenever you\nneed to talk,\nwe‘re open\n\n[—] text eseses\n[J] KidsHelpPhone.ca\n\n(@, call 1—800—663—6868 Kids Help Phone ©"
}	
}