rapidocr v3.4.0发布了

更新日志

(issue #530)支持PaddleOCR在v3.2.0中发布的PP-OCRv5 英文/泰文/希腊文三个文本识别模型

这三个语种默认对应LangRec.EN、LangRec.TH和LangRec.EL。具体可参见官方文档中模型列表一节。

下面给出希腊文识别示例：

from rapidocr import LangRec, ModelType, OCRVersion, RapidOCRengine = RapidOCR(params={"Rec.lang_type": LangRec.EL,"Rec.model_type": ModelType.MOBILE,"Rec.ocr_version": OCRVersion.PPOCRV5,}
)
img_url = "https://github.com/RapidAI/RapidOCR/blob/main/python/tests/test_files/el_rec.jpg?raw=true"
result = engine(img_url, use_det=False, use_cls=False, use_rec=True)
print(result)result.vis("vis_result.jpg")

(issue #531) 添加日志打印开关

from rapidocr import RapidOCRengine = RapidOCR(params={"Global.log_level": "critical"})  # 注意这里改为critical，就不会打印日志了img_url = "https://img1.baidu.com/it/u=3619974146,1266987475&fm=253&fmt=auto&app=138&f=JPEG?w=500&h=516"
result = engine(img_url)
print(result)result.vis("vis_result.jpg")
print(result.to_markdown())

(issue #528) 修复保存中文路径乱码问题

(issue #527) 修复PyTorch>=2.6时，加载模型权重问题

(issue #544) 统一det、cls和rec各个阶段统计运行时间的范围

都统一改为从前处理、模型推理和后处理都算上。

重构单元测试部分

主要是拆分原有的test_main.py文件，利于后续维护

---

🚀 Features

• add en/el/th ppocr_v5_rec_mobile paddle/onnx/openvino (#546) by @SWHL in 017b8fd

🐛 Bug Fixes

• fixed issue #544 (#545) by @SWHL in a40454c
• remove deprecated getLevalName by @SWHL in 386c415
• fixed issue #531 by @SWHL in 336e642
• fixed issue #528 by @SWHL in a0bb926
• fixed issue #527 by @SWHL in af65796

📚 Documentation

• update logo to v2 by @SWHL in e4afa55
• add QQ group7 by @SWHL in cfebc47

⚙️ Miscellaneous Tasks

• fixed the error of running unit testing code by @SWHL in bb07136
• update ucloud url by @SWHL in 66d5a04

🎉 Contributors

• @SWHL

Full Changelog: 3.4.0