tencentcloud-ocr-generalaccurate

Tencent Cloud General Text Recognition (High-Precision Edition) (GeneralAccurateOCR)

Purpose

Call the Tencent Cloud OCR General Text Recognition (High-Precision Edition) API to accurately extract text from images.

Core capabilities:

Text recognition: High-precision recognition of text content in images, returning the complete recognized text
PDF support: Supports text recognition for PDF files (single page)
Single-character information: Optionally returns the position and confidence information of each single character
Multilingual resume structured recognition: Based on OCR results, performs structured extraction and formatted output for resumes (see references/resume-parsing.md）

Official documentation:https://cloud.tencent.com/document/api/866/37831

📚 Available resources

References (scenario-based guidance)

references/resume-parsing.md - Multilingual resume structured recognition guide (processing flow, Prompt templates, output formatting templates, formatting rules)

When to use

This skill is triggered when the user raises the following needs:

Need to extract text content from images or PDFs
Need to perform text recognition on various documents, images, etc.
Any scenario involving general text OCR recognition
Need to recognize and structurally extract resume information from resume images/PDFs (please refer to references/resume-parsing.md guide)
Need to recognize and format multilingual resumes for output (please refer to references/resume-parsing.md guide)

Environment requirements

Python 3.6+
Dependency:tencentcloud-sdk-python(via pip install tencentcloud-sdk-python installation)
Environment variables:
- TENCENTCLOUD_SECRET_ID: Tencent Cloud API key ID
- TENCENTCLOUD_SECRET_KEY: Tencent Cloud API key Key

Usage

Run the scripts/main.py script to complete text recognition.

Request parameters

Parameter	Type	Required	Description
ImageBase64	str	No (choose one of two)	Image Base64 value, no more than 10MB
ImageUrl	str	No (choose one of two)	Image URL address, preferred
IsPdf	bool	No	Whether to enable PDF recognition, default false
PdfPageNumber	int	No	PDF page number to recognize, valid when IsPdf is true, default 1
IsWords	bool	No	Whether to return single-character information, default false
UserAgent	str	No	Request source identifier (optional), used to track the call source, uniformly fixed as`Skills`

⚠️ UserAgent parameter usage guide

--user-agentThe parameter is optional, uniformly fixed asSkills, no manual passing required. Used to identify the API call source for tracking and statistics:

Calling framework	--user-agent parameter value	Description
All frameworks	`Skills`	Uniform fixed value, also defaults to this value when not passed

Implementation Notes：

via--user-agentpassed through command-line arguments, the SDK will concatenate them as SDK_PYTHON_x.x.x; Skills injected into the request
uniformly fixed toSkills, and defaults to this value when not passed
this identifier will be recorded in the ES log's ReqBody.RequestClient field, which can be used to trace the source

Output Format

Return a JSON format result after successful recognition:

{
  "raw_text": "识别到的完整文字内容\n第二行文字\n第三行文字",
  "RequestId": "xxx"
}

Return when there is no text:

{
  "raw_text": "",
  "message": "No text detected in the image.",
  "RequestId": "xxx"
}

Usage Example

# 基础调用示例（--user-agent 默认为 Skills，可不传）
python scripts/main.py --image-url "https://example.com/document.jpg"

# 使用 Base64 文件调用
python scripts/main.py --image-base64 "/path/to/document.jpg"

# 识别 PDF 文件中的文字
python scripts/main.py --image-url "https://example.com/doc.pdf" \
  --is-pdf true --pdf-page-number 1

# 返回单字信息
python scripts/main.py --image-url "https://example.com/document.jpg" --is-words true

Key Configuration

Step 1: Obtain/Purchase OCR Service

🔗 Tencent Cloud Optical Character Recognition OCR Purchase Page

On the purchase page, select General Text Recognition (High Precision Edition) to complete the purchase.

Step 2: Obtain API Key

🔗 Tencent Cloud API Key Management

Step 3: Set Environment Variables

Linux / macOS：

export TENCENTCLOUD_SECRET_ID="你的SecretId"
export TENCENTCLOUD_SECRET_KEY="你的SecretKey"

Windows (PowerShell)：

$env:TENCENTCLOUD_SECRET_ID = "你的SecretId"
$env:TENCENTCLOUD_SECRET_KEY = "你的SecretKey"