不用第三方 API!FastAPI + PaddleOCR 自建身份证 OCR 服务实战

35 阅读3分钟

1. 环境构建

1.1 新建fastapi项目

利用PyCharm新建一个FastAPI项目

1.2 添加额外需要的依赖

创建好项目后添加一些其他的依赖 1)图片识别需要:

pip install paddlepaddle pip install paddleocr

2)上传图片需要:

pip install python-multipart

1.3 项目环境

利用 pip list 查看相关的依赖版本

Package               Version
--------------------- -----------
aistudio-sdk          0.3.8
annotated-doc         0.0.4
annotated-types       0.7.0
anyio                 4.12.1
bce-python-sdk        0.9.59
certifi               2026.1.4
chardet               5.2.0
charset-normalizer    3.4.4
click                 8.3.1
colorama              0.4.6
colorlog              6.10.1
fastapi               0.128.0
filelock              3.20.2
fsspec                2025.12.0
future                1.0.0
h11                   0.16.0
hf-xet                1.2.0
httpcore              1.0.9
httptools             0.7.1
httpx                 0.28.1
huggingface_hub       1.2.4
idna                  3.11
imagesize             1.4.1
modelscope            1.33.0
networkx              3.6.1
numpy                 2.4.0
opencv-contrib-python 4.10.0.84
opt-einsum            3.3.0
packaging             25.0
paddleocr             3.3.2
paddlepaddle          3.2.2
paddlex               3.3.12
pandas                2.3.3
pillow                12.1.0
pip                   23.2.1
prettytable           3.17.0
protobuf              6.33.2
psutil                7.2.1
py-cpuinfo            9.0.0
pyclipper             1.4.0
pycryptodome          3.23.0
pydantic              2.12.5
pydantic_core         2.41.5
pypdfium2             5.3.0
python-bidi           0.6.7
python-dateutil       2.9.0.post0
python-dotenv         1.2.1
python-multipart      0.0.21
pytz                  2025.2
PyYAML                6.0.2
requests              2.32.5
ruamel.yaml           0.19.1
safetensors           0.7.0
setuptools            80.9.0
shapely               2.1.2
shellingham           1.5.4
six                   1.17.0
starlette             0.50.0
tqdm                  4.67.1
typer-slim            0.21.1
typing_extensions     4.15.0
typing-inspection     0.4.2
tzdata                2025.3
ujson                 5.11.0
urllib3               2.6.3
uvicorn               0.40.0
watchfiles            1.1.1
wcwidth               0.2.14
websockets            15.0.1

2. 代码实现

from fastapi import FastAPI, UploadFile, File
from paddleocr import PaddleOCR
import uuid
import os
import re

app = FastAPI()

# 建议:OCR 实例全局初始化
ocr = PaddleOCR(lang='ch')
# 存放图片的文件夹
UPLOAD_DIR = "temp_images"
os.makedirs(UPLOAD_DIR, exist_ok=True)


@app.get("/")
async def root():
    return {"message": "Hello World"}


@app.post("/ocr/idcard")
async def ocr_idcard(file: UploadFile = File(...)):
    """
    接收身份证图片并进行OCR识别,有一定的识别失败率。
    """
    # 1. 保存上传的图片
    suffix = os.path.splitext(file.filename)[-1]
    temp_filename = f"{uuid.uuid4()}{suffix}"
    temp_path = os.path.join(UPLOAD_DIR, temp_filename)

    with open(temp_path, "wb") as f:
        f.write(await file.read())
    try:
        # OCR识别
        ocr_result = ocr.predict(temp_path)
        data = ocr_result[0]
        # 对于不同版本的paddleocr可能存在识别出来的数据不一样。
        texts = data.get("rec_texts", [])

        # 解析身份证
        idcard = parse_idcard(texts)

        id_number = idcard.get("id_number")

        # 判断身份证是否合法
        if not id_number or not is_valid_id_number(id_number):
            return {
                "success": False,
                "message": "图片有误,请重新上传身份证照片"
            }

        return {
            "success": True,
            "message": "成功",
            "texts": texts,
            "result": idcard
        }

    finally:
        # 最后删除临时文件夹的图片,看需求决定是否保留
        if os.path.exists(temp_path):
            os.remove(temp_path)


def parse_idcard(texts: list[str]):
    """
    从OCR文本中解析出身份证和姓名等信息
    """
    result = {
        "name": None,
        "gender": None,
        "birthday": None,
        "address": None,
        "id_number": None
    }

    i = 0
    address_lines = []

    while i < len(texts):
        text = texts[i].replace(" ", "").strip()
        if text.startswith("姓名"):
            value = text.replace("姓名", "")
            if value:
                result["name"] = value
            else:
                # 下一行是姓名
                if i + 1 < len(texts):
                    result["name"] = texts[i + 1].strip()
        elif text.startswith("性别"):
            value = text.replace("性别", "")
            if value:
                result["gender"] = value[0]
            else:
                if i + 1 < len(texts):
                    result["gender"] = texts[i + 1].strip()
        elif text.startswith("出生"):
            value = text.replace("出生", "")
            if value:
                result["birthday"] = value
            else:
                if i + 1 < len(texts):
                    result["birthday"] = texts[i + 1].strip()

        # 住址(可能多行)
        elif text.startswith("住址"):
            value = text.replace("住址", "")
            if value:
                address_lines.append(value)

            # 向后继续收集,直到遇到“公民身份号码”
            j = i + 1
            while j < len(texts):
                next_text = texts[j].replace(" ", "").strip()
                if "公民身份号码" in next_text:
                    break
                address_lines.append(texts[j].strip())
                j += 1

            result["address"] = "".join(address_lines)
        # 身份证号(优先关键字,其次正则)
        elif "公民身份号码" in text:
            value = text.replace("公民身份号码", "")
            if value:
                result["id_number"] = value
            else:
                if i + 1 < len(texts):
                    result["id_number"] = texts[i + 1].strip()

        # 防止 OCR 把关键字漏了
        else:
            m = re.search(r"\d{17}[\dXx]", text)
            if m:
                result["id_number"] = m.group()
        i += 1
    return result


def is_valid_id_number(id_number: str) -> bool:
    """
    简单判断是否为18位身份证号
    """
    if not id_number:
        return False

    id_number = id_number.upper().replace(" ", "")

    # 17位数字 + 1位数字或X
    return bool(re.fullmatch(r"\d{17}[\dX]", id_number))


3. 测试接口

启动项目后利用接口调用工具测试一下接口,这里使用的是apifox

最终输出结果如下:

{
  "success": true,
  "message": "成功",
  "texts": [
    "姓名xx",
    "性别xx",
    "民族xx",
    "出生xxx",
    "住址xxxx",
    "xxx",
    "xx",
    "61xxxx"
  ],
  "result": {
    "name": "xxx",
    "gender": "xx",
    "birthday": "xx",
    "address": "xxxx",
    "id_number": "xxxxxxxxx"
  }
}

4.更多信息

关于更多的信息请查看框架的官网:

fastapi : fastapi.tiangolo.com/

pp飞桨:www.paddlepaddle.org.cn/

PaddleOCR: github.com/PaddlePaddl…