Baidu OCR icon

Baidu OCR

使用百度智能云 OCR API 识别图片和文档中的文字

Actions13

身份证识别

AI-generated

Summary

Extract identity card information from image or PDF.

Inputs

  • inputType — 选择输入方式：binaryData（二进制数据）或 url（图片网络地址）
  • binaryPropertyName — 包含图片或PDF的二进制属性名，支持多个属性名（用逗号分隔），节点会自动检测文件格式
  • pdfPageRange — PDF页码范围，如 all、1、1-5、1,3,5
  • imageUrl — 图片的网络地址
  • idCardSide — 选择要识别身份证的哪一面，正面或反面

Output shape

an array of result items, each containing a JSON object with extracted text and OCR result details

Each item's json field holds the OCR output; when continueOnFail is enabled, errors are returned as {error: message}.

Examples

Example 1: Process a binary image uploaded by a previous node to recognize the front side of an ID card

inputType=binaryData, binaryPropertyName=data0, idCardSide=front

Example 2: OCR a publicly accessible image URL to extract both sides of an ID card

inputType=url, imageUrl=https://example.com/id.jpg, idCardSide=front

Example 3: Identify pages 2‑4 of a multi‑page PDF containing an ID card

inputType=binaryData, binaryPropertyName=data0, pdfPageRange=2-4

Links

Discussion