Image OCR
Image OCR uses optical character recognition to prepare readable text from an image or scanned document. In iDialogue, OCR is a focused capability within Intelligent Document Processing, not the end of the workflow.
Once text is prepared, a configured agent can summarize it, answer questions, extract the intended values, or use permitted Salesforce tools according to business and human-review rules.
Common file examples
- Business cards attached to Lead, Contact, Account, or campaign workflows.
- Receipts and expense images.
- Scanned contracts, agreements, applications, invoices, and forms.
- Photographs containing relevant business text.
- Multi-page PDFs that require readable text before analysis.
- Supported HEIC or HEIF images that require conversion before processing.
Typical workflow
- A user attaches an image or document to the intended Salesforce record.
- Salesforce automation sends the File identifiers and related record context to iDialogue.
- iDialogue converts the image when necessary and prepares readable text.
- The configured agent receives the file content and related Salesforce context.
- The agent follows extraction, duplicate-prevention, record-action, and human-review rules.
OCR prepares text; it does not guarantee that every image is legible or that every extracted business value is correct. Use representative files and human review for consequential actions.