# paddleOCR

倉庫：[https://github.com/PaddlePaddle/PaddleOCR?tab=readme-ov-file](https://github.com/PaddlePaddle/PaddleOCR?tab=readme-ov-file)

文檔：[https://paddlepaddle.github.io/PaddleOCR/main/quick_start.html#2-paddleocr](https://paddlepaddle.github.io/PaddleOCR/main/quick_start.html#2-paddleocr)

# 環(huán)境安裝

```shell

python -m pip install paddlepaddle-gpu==3.0.0b1 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/

pip install paddleocr

```

# 示例代碼```pythonfrom paddleocr import PaddleOCR, draw_ocr# Paddleocr supports Chinese, English, French, German, Korean and Japanese# You can set the parameter `lang` as `ch`, `en`, `french`, `german`, `korean`, `japan`# to switch the language model in orderocr = PaddleOCR(use_angle_cls=True, lang='en') # need to run only once to download and load model into memoryimg_path = 'PaddleOCR/doc/imgs_en/img_12.jpg'result = ocr.ocr(img_path, cls=True)for idx in range(len(result)): res = result[idx] for line in res: print(line)# draw resultfrom PIL import Imageresult = result[0]image = Image.open(img_path).convert('RGB')boxes = [line[0] for line in result]txts = [line[1][0] for line in result]scores = [line[1][1] for line in result]im_show = draw_ocr(image, boxes, txts, scores, font_path='/path/to/PaddleOCR/doc/fonts/simfang.ttf')im_show = Image.fromarray(im_show)im_show.save('result.jpg')```# 更換高精度模型PP-OCRv4_server```Linuxwget https://paddleocr.bj.bcebos.com/PP-OCRv4/chinese/ch_PP-OCRv4_det_server_infer.tarwget https://paddleocr.bj.bcebos.com/PP-OCRv4/chinese/ch_PP-OCRv4_rec_server_infer.tarocr = PaddleOCR(use_angle_cls=False, lang='ch', det_model_dir=det_model_dir, rec_model_dir=rec_model_dir)```# 問題DEBUG## [更換了高精度的檢測模型det和識別模型rec以后，cuda會報錯: [Hint: 'CUDNN_STATUS_NOT_SUPPORTED'. The functionality requested is not presently supported by cuDNN. ] ](https://github.com/PaddlePaddle/PaddleOCR/issues/10760)實(shí)測ubuntu20.04+cuda11.8+cudnn8.6可以運(yùn)行server模型paddlepaddle-gpu 2.5.1, cuda 11.6 + cudnn8.6.0 測試可以環(huán)境重新安裝文檔：[https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/install/pip/linux-pip.html](https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/install/pip/linux-pip.html)## 查看cudnn版本```Ccat /usr/local/cuda-11.8/include/cudnn_version.h// 得到如下代碼#define CUDNN_MAJOR 8#define CUDNN_MINOR 9#define CUDNN_PATCHLEVEL 3// CUDNN_MAJOR、CUDNN_MINOR 和 CUDNN_PATCHLEVEL 分別表示主版本號、次版本號和修補(bǔ)版本號。這些定義表明當(dāng)前庫的版本是 8.9.3。```## 安裝cudnn參考教程：[https://blog.csdn.net/weixin_73728875/article/details/129054053](https://blog.csdn.net/weixin_73728875/article/details/129054053)```Linuxtar -xvf cudnn-linux-*****#TAB鍵可以快速補(bǔ)全文件名# 復(fù)制cudnn相關(guān)文件到cuda目錄下sudo cp include/* /usr/local/cuda-11.8/include/sudo cp lib/* /usr/local/cuda-11.8/lib64/# 添加可執(zhí)行權(quán)限sudo chmod +x /usr/local/cuda-11.8/include/cudnn.hsudo chmod +x /usr/local/cuda-11.8/lib64/libcudnn*```本文由[mdnice](https://mdnice.com/?platform=6)多平臺發(fā)布