Baidu OCRInstall

Summary

Recognize general machine-printed invoices from binary image or PDF data or from an image URL, optionally on selected PDF pages.

Inputs

• inputType — Source of input: "binaryData" for binary data from upstream nodes or "url" for an online image URL.
• binaryPropertyName — Comma-separated list of binary property names that may contain the image or PDF data; the node detects which property holds the file.
• pdfPageRange — PDF page range to process (e.g., "all", "1-5", "1,3,5"); ignored for image inputs.
• imageUrl — Direct URL of an image to be processed when inputType is set to "url".

Output shape

array of objects each containing a json field with OCR extracted text/structured results and a pairedItem field indicating the input item index

When continueOnFail is enabled, failed items may return an object with an error field instead of OCR data.

Examples

Example 1: Process binary image data from upstream node

inputType = binaryData, binaryPropertyName = data, default pdfPageRange = all

Example 2: Process image from a URL

inputType = url, imageUrl = https://example.com/invoice.jpg

Links

• n8n community nodes installation guide