正确安装PaddleOCR的方法
paddleocr安装起来太费劲了
pip install patch-ng
pip install paddleocr --use-pep517 (失败)
1.首先去github下载:https://github.com/PaddlePaddle/PaddleOCR,然后找个文件夹解压。(从CSDN中点开链接时会提供加速功能,跳转到了https://gitcode.net/mirrors/paddlepaddle/paddleocr?utm_source=csdn_github_accelerator )
2.安装PaddlePaddle
按照官网提供的命令:
python -m pip install paddlepaddle==2.5.1 -i https://mirror.baidu.com/pypi/simple
3 cd到刚才解压的PaddleOCR文件夹中,执行:
pip install -r requirements.txt
红色字提示:ERROR: No matching distribution found for opencv-python==4.2.0.32
修改 requirements.txt 直接去掉 ==4.2.0.32
再次执行:pip install -r requirements.txt
执行:pip install paddleocr
在执行到安装PyPDF时出错,提示:error: command 'swig.exe' failed: None
执行:pip install swig
再次执行:pip install paddleocr 终于成功了
2.代码
from paddleocr import PaddleOCR, draw_ocr ocr = PaddleOCR(use_angle_cls=True, use_gpu=False) img_path = r'F:\1.png' result = ocr.ocr(img_path, cls=True) for line in result: print(line)
首次在CMD中运行代码时,下载了3 个文件到 C:\Users\用户名\.paddleocr\whl\rec\ch\ch_PP-OCRv4_rec_infer 目录,分别是:inference.pdiparams inference.pdiparams.info inference.pdmodel
删除掉下载后解压的paddleocr-develop目录,在IDLE中再次执行代码,IDLE中输出结果正常,但是有黑框一闪。
参考:https://blog.csdn.net/weixin_44063045/article/details/129211758
https://blog.51cto.com/u_15671528/5974388
https://pythonjishu.com/ustjibibiytohrb/
https://blog.csdn.net/weixin_64974855/article/details/132768887
https://www.paddlepaddle.org.cn/
https://blog.csdn.net/weixin_42708301/article/details/119864744
https://blog.csdn.net/qq_44133071/article/details/132521396