/ocr | Type: Embedded | PCID required: No
Extract text from images and PDF files using OCR. Supports 30+ languages, synchronous and asynchronous processing, and detailed layout analysis with bounding boxes.
Tools
| Tool | Description |
|---|---|
ocr_process | Extract text from images and PDFs |
ocr_process
Extract text from one or more images or PDF files using Optical Character Recognition. Parameters:| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
fileUrls | string[] | Yes | — | URLs of the image or PDF files to process |
languageHints | enum[] | No | — | Language hints to improve OCR accuracy: "eng", "spa", "fra", "deu", "ita", "por", "rus", "chi_sim", "chi_tra", "jpn", "kor", "nld", "pol", "tur", "ara", "hin", "tha", "vie", "ukr", "ces", "hun", "ron", "bul", "hrv", "slk", "slv", "lit", "lav", "est", "fin", "swe", "nor", "dan", "ell", "heb" |
extractTextOnly | boolean | No | true | When true, returns only extracted text. When false, includes detailed layout analysis with bounding boxes. |
collectionId | string | No | — | Collection ID to store the OCR results |
async | boolean | No | false | Whether to process asynchronously |
| Field | Type | Description |
|---|---|---|
artifactId | string | ID of the OCR result artifact |
artifactName | string | Name of the OCR result artifact |
artifactUrl | string | URL to access the OCR results |
mimeType | string | MIME type of the result |