跳转到主要内容

项目集成paddleocr功能

示例代码下载

参考网站

  • https://www.paddlepaddle.org.cn/
  • https://github.com/PaddlePaddle/PaddleOCR
  • https://fastapi.tiangolo.com/

安装paddle和paddle ocr环境

# 安装paddlepaddle
# https://www.paddlepaddle.org.cn/
pip install paddlepaddle==2.6.1 -i https://pypi.tuna.tsinghua.edu.cn/simple

# 安装paddleocr
# 提示，截止至2024年5月2日，需要的python版本为3.11.更高版本会出现安装不上的情况
# 提示，截止至2024年5月2日，paddleocr==2.7.5有bug
# 需要修改的地方
#  C:\Python311\Lib\site-packages\paddleocr\paddleocr.py
# 第54行之后
# from ppstructure import predict_system 


pip install paddleocr==2.7.5 -i https://mirror.baidu.com/pypi/simple


# 可能出现其他安装失败情况，请使用清华源安装,更新pip工具
pip install --upgrade pip wheel setuptools

安装fast api服务框架

# 安装fastAPI
pip install fastapi -i https://pypi.tuna.tsinghua.edu.cn/simple

# 安装web容器
pip install "uvicorn[standard]" -i https://pypi.tuna.tsinghua.edu.cn/simple

# 安装from参数表单
pip install python-multipart -i https://pypi.tuna.tsinghua.edu.cn/simple

编写后台代码

import paddle as paddle
from paddleocr import PaddleOCR, draw_ocr
from PIL import Image
from fastapi import FastAPI, File, UploadFile
from fastapi.responses import HTMLResponse
from fastapi.staticfiles import StaticFiles
import time
import uvicorn

# 打印paddle版本
print(paddle.__version__)

app = FastAPI()


def get_image_ocr(img_path: str):
    # 创建一个OCR实例
    ocr = PaddleOCR(
        use_angle_cls=True, lang="ch"
    )  # need to run only once to download and load model into memory
    # img_path = "./images/test.png"
    # 使用PaddleOCR进行文字检测和识别
    ocr_result = ocr.ocr(img_path, cls=True)
    # 显示结果
    for line in ocr_result:
        print(line)

    # 对现实结果在原图进行标注
    result = ocr_result[0]
    image = Image.open(img_path).convert("RGB")
    boxes = [elements[0] for elements in result]
    # pairs = [elements[1] for elements in ocr_result]
    # txts = [pair[0] for pair in pairs]
    # scores = [pair[1] for pair in pairs]
    im_show = draw_ocr(image, boxes)
    im_show = Image.fromarray(im_show)
    im_show.save(f"{img_path}")
    # im_show.save("./images/result.jpg")
    return {"image": img_path, "result": result}


app.mount("/static", StaticFiles(directory="./static"), name="static")


@app.get("/")
def read_root():
    html_content = """
    <html>
        <head>
            <title>OCR</title>
        </head>
        <body>
            <h1>OCR</h1>
            <p>Upload an image to extract text from it.</p>
        </body>
    </html>
    """
    return HTMLResponse(content=html_content, status_code=200)


sequence_counter = 1


@app.post("/ocr/")
async def ocr_image(file: UploadFile | None = File(None)):
    # 判断文件是否存在
    if file is None:
        return {"code": 400, "message": "No file found", data: None}

    # 生成时间戳保存上传图片以免被覆盖
    current_time = int(time.time() * 10000)
    unique_sequnce = current_time + sequence_counter
    new_fileName = f"./static/{unique_sequnce}--{file.filename}.jpg"
    # 将文件写入磁盘
    with open(f"{new_fileName}", "wb+") as fsio:
        fsio.write(await file.read())
    pass
    # 调用OCR方法
    ocr_result = get_image_ocr(new_fileName)

    return {"code": 200, "message": "ok", "data": ocr_result}


# 启动服务
uvicorn.run(app, host="0.0.0.0", port=8000)
回到顶部