HyperLPR3 车牌识别

Linux 之前安装了python3

apt install python3.8-venv
cd /root
python3 -m venv HyperLPR3
REM cd HyperLPR3
source HyperLPR3/bin/activate
参考 https://www.jb51.net/article/222885.htm
python -m pip install hyperlpr3
里面有fastapi，opencv等

错误：
ERROR: Could not find a version that satisfies the requirement pydantic!=1.8,!=1.8.1,!=2.0.0,!=2.0.1,!=2.1.0,<3.0.0,>=1.7.4 (from fastapi->hyperlpr3) (from versions: none)
ERROR: Could not find a version that satisfies the requirement charset-normalizer<4,>=2 (from requests->hyperlpr3) (from versions: none)
ERROR: No matching distribution found for charset-normalizer<4,>=2 (from requests->hyperlpr3)
ERROR: Could not find a version that satisfies the requirement numpy>=1.21.6 (from onnxruntime->hyperlpr3) (from versions: none)
ERROR: No matching distribution found for numpy>=1.21.6 (from onnxruntime->hyperlpr3)

解决办法，可能是网络问题，也可能是pip版本比较低
pip install --upgrade pip
    Successfully installed pip-24.0

启动WebAPI

nohup lpr3 rest --port 8715 --host 0.0.0.0 &

浏览器测试 http://192.168.1.230:8715/api/v1/docs

如果上面的网页（上海OK，武汉空白页面）出现空白，浏览器发现swagger js文件加载有问题。

那么修改hosts文件加入

# fastly.jsdelivr.net
 151.101.1.229  cdn.jsdelivr.net

参考：

GitHub - szad670401/HyperLPR: 基于深度学习高性能中文车牌识别 High Performance Chinese License Plate Recognition Framework.