Use the GLM-4V-Flash vision model to recognize text in images.
Installation
git clone https://github.com/Wcowin/zhipu-OCR-skill.git
cd zhipu-OCR-skill
pip install -r requirements.txt
Usage
python3 scripts/ocr.py <图片路径> [提示词]
<图片路径>: Required, image file path[提示词]: Optional, default "Please recognize all text in the image and output it completely"-v: Show detailed output
Configuration
Set the environment variable ZHIPUAI_API_KEY(obtained from https://open.bigmodel.cn/usercenter/apikeys ).
Supported Formats
Common image formats such as JPG, PNG, GIF, BMP, WebP.
Use Cases
- The user sends an image and requests text recognition
- Screenshot text extraction
- Scanned document OCR
- Table content recognition
- Handwritten text recognition
Notes
- A valid Zhipu AI API Key is required
- A network connection is required to access the API