用HyperLPR创作Python车牌识别项目

用HyperLPR创作Python车牌识别项目

用HyperLPR创作python车牌识别项目

HyperLRP是一个开源的、基于深度学习高性能中文车牌识别库，由北京智云视图科技有限公司开发，支持PHP、C/C++、Python语言，Windows/Mac/Linux/Android/IOS 平台。进行一个车牌识别的入门探索。

1、搭建环境：
1.1在pycharm里创建工程HyperLPR,选择python3.6解释器
1.2在pycharm中找到并点击下方图片中的Terminal安装必备包

1.3安装必备包

pip install keras==2.0.9 -i  --trusted-host pypi.doubanpip install Theano -i  --trusted-host pypi.doubanpip install numpy==1.16.0 -i  --trusted-host pypi.doubanpip install Scipy -i  --trusted-host pypi.doubanpip install opencv-python==3.4.3.18 -i  --trusted-host pypi.doubanpip install scikit-image -i  --trusted-host pypi.doubanpip install pillow -i  --trusted-host pypi.doubanpip install tensorflow==1.2.0 -i  --trusted-host pypi.doubanpip install h5py

以上所有包都安装指定版本

2、下载HyperLPR包
链接：
提取码：chy1
里面其实只需要三个文件：hyperlpr_py3，model，Font
删掉hyperlpr_py3中_pycache_文件(不需要)
把以上三个文件全部导入到项目工程即可
导入后如下图

3、接着把每个py文件里的from . . 删掉
4、最后在main函数里测试一下
图片为jpg或png都可以
main.py

import pipline as pp
import cv2image = cv2.imread("D:/IECS_PHOTOS/1.jpg")
image, res = pp.SimpleRecognizePlate(image)
print(res)

Using TensorFlow backend.
2021-04-26 17:21:58.849879: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use SSE instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.850290: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use SSE2 instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.850602: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.850913: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.851263: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.851574: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.851884: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2021-04-26 17:21:58.852227: W c:tf_jenkinshomeworkspacerelease-winmwindowspy36tensorflowcoreplatformcpu_feature_guard:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
(1048, 1680, 3)
校正角度 h  0 v 90
keras_predict [0.10701188 0.9172991 ]
c769b466
e2e: ('京KBT355', 0.991141106401171)
校正 0.6603515148162842 s
分割 0.05764651298522949
410
寻找寻找最佳点 0.0199129581451416
字符识别 0.08989286422729492
分割和识别 0.17144370079040527 s
车牌: 京KBT355 置信度: 0.9851458583559308
3.277062177658081 s
['京KBT355']Process finished with exit code 0

完成！