ceven

OCR Web Service

Extracts text and word coordinates from images or documents and pipes that data into your downstream database or CRM for automated processing.

Try OCR Web Service in Ceven

Ask Ceven anything
Standard

Why use Ceven?

  1. AI native OCR Web Service integration

    • Describe the outcome and Ceven picks the right OCR Web Service calls, fills the parameters, and checks the result.
    • Structured, agent friendly tool schemas so each call runs reliably instead of by guesswork.
    • Rich coverage for reading, writing, and querying your OCR Web Service data, across all 4 of its actions.

  2. Managed auth

    • Built in OAuth with automatic token refresh and rotation.
    • One place to manage, scope, and revoke OCR Web Service access.
    • Per user and per environment credentials instead of shared keys.

  3. Agent optimized design

    • Actions are tuned from real success and error rates so reliability climbs over time.
    • Full execution logs so you always know what ran in OCR Web Service, when, and on whose behalf.
    • The agent pauses and asks when OCR Web Service is unclear instead of plowing ahead.

  4. Enterprise grade security

    • Fine grained access so you control which agents and people can reach OCR Web Service.
    • Least privilege by default, read scopes first and only the writes a workflow needs.
    • A full audit trail of every OCR Web Service action to support review and sign off.

Supported tools

Every action Ceven's agents can run on OCR Web Service, and when to use it.

Recognize text
Use this when you need to perform OCR on an image to retrieve raw text, output documents, and word coordinates.
Get account information
Pull current account details including remaining pages, subscription plan, and expiration date to check capacity.
Retrieve processing logs
Pull OCR processing logs for a specific date range to troubleshoot failed document conversions.
Get account credentials
Use this to extract credentials from connection metadata before invoking other service operations.
Extract invoice text
Perform a recognize operation specifically on invoice images to pull out line items and totals.
Verify page balance
Check the remaining page count to ensure there are enough credits for a bulk upload job.
Audit OCR errors
Pull the log for a specific day to identify which documents failed the recognition process.
Get subscription status
Check if the OCR Web Service plan is active or nearing expiration to prevent workflow downtime.
Parse document coordinates
Use the recognize tool to find the exact x and y position of words for template based extraction.
Sync account metadata
Pull the latest account credentials and plan details to update the internal workflow state.
Batch process images
Iterate through a list of image URLs and call the recognize operation for each one.
Check API expiration
Pull the account expiration date to trigger a renewal alert in the management dashboard.
OCR Web Service Log
Tool to retrieve ocr processing logs. use after specifying date range for your account.
OCRWebService Recognize
Tool to call soap recognize operation. use when performing ocr on an image to retrieve text, output document, word coordinates, and errors.

14 actions · scroll to see them all

Frequently asked questions

Ceven monitors your remaining page count by calling the Get Account Information tool before initiating large batch jobs. If your balance is too low to complete the requested set of documents, the agent will pause the workflow and send a notification to the administrator instead of allowing the API to return a credit error mid process. This prevents partial data imports and ensures that critical documents are not skipped during a high volume run. You can set up a custom alert in Ceven to notify you when your OCR Web Service pages drop below a specific threshold.
Ceven uses the recognize operation provided by OCR Web Service to extract text. The accuracy for handwritten text depends on the clarity of the image and the specific engine version being used by the API. For printed text the accuracy is very high, but for cursive or messy handwriting the output may contain errors. We recommend using a verification step in your workflow where the agent compares the extracted text against known patterns or sends the result to a human for a quick check if the confidence score is low.
When a recognize call fails, Ceven uses the OCR Web Service Log tool to pull the specific error code associated with that transaction. The agent then analyzes the log to determine if the failure was due to a file format issue, a timeout, or an account limit. If the error is transient, the agent can automatically retry the request. If the error is permanent, such as an unsupported file type, the agent marks the record as failed in your database and attaches the log entry for your review.
Yes, OCR Web Service has specific limits on the size and resolution of images processed through their API. If you upload a file that exceeds these limits, the API will return an error. To handle this, Ceven can be configured to check file size before sending the request. If a file is too large, the agent can attempt to compress the image or split a multi page PDF into smaller chunks before calling the recognize operation to ensure the request is accepted by the server.
Ceven acts as a secure bridge between your storage and the OCR Web Service API. Files are transmitted over encrypted channels using HTTPS. We do not store your documents on our own servers long term; we simply pass the image to the API, receive the text response, and push that data to your destination system. Because we use the account credentials stored in your secure connection metadata, your data remains isolated to your specific account and is never shared across different customer workflows.
OCR Web Service supports both protocols, but Ceven abstracts this complexity away from the user. Depending on the specific tool being called, the agent uses the most efficient method available. For simple text recognition, the REST API is typically used for speed and lower overhead. For more complex operations that require detailed SOAP envelopes, Ceven handles the XML construction and parsing automatically so you only interact with simple text and JSON objects in your workflow builder.
The most frequent issue is the exhaustion of the monthly page quota, which is a hard limit tied to the subscription tier. When this happens, the API returns a specific error indicating that the page limit has been reached. Because this is a tier gated limitation, the only way to resolve it is to upgrade the plan or wait for the billing cycle to reset. Ceven helps mitigate this by proactively checking the account information tool so you know exactly when you are running low on credits.
OCR Web Service provides the raw text and the coordinates of every word found on the page. It does not natively know what a total or a date is. However, Ceven uses the coordinate data and the raw text output to perform field extraction. The agent looks for keywords like Total or Date and then captures the text located immediately to the right or below those words. This allows the agent to turn an unstructured text dump into a structured record for your CRM.

Alternatives to OCR Web Service

Other tools that solve a similar problem. Ceven supports these too, so you can switch or run more than one at once.

Tesseract logoTesseractABBYY logoABBYYGoogle Cloud Vision logoGoogle Cloud Vision

Try Ceven on your stack

Plug Ceven on top of the tools you already run. Connect OCR Web Service and the rest of your stack, describe the outcome, and its agents handle the work end to end, days of it in minutes.

Get started for free