简介
- OCR(Optical Character Recognition)全称光学字符识别, 通俗的讲就是计算机识别图像上面的文字并且提取出来。这对提取运算速度以及识别准确率都有很高的要求。
- 两个项目均CRNN网络结构
- chineseocr_lite运行方式简单,PaddleOCR自定义功能强
- 笔者运行环境:Anaconda3的Python3.7 完美运行两个项目
chineseocr_lite
pip install -r ./requirements.txt -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
python backend/main.py
- 项目运行成功:
- PC界面:
- Android界面:
PaddleOCR
pip install paddlepaddle==2.0.0rc1 -i https://mirror.baidu.com/pypi/simple
pip install paddle
pip3 install paddlehub --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
检测模型:.\PaddleOCR_dygraph\deploy\hubserving\ocr_det
识别模型:.\PaddleOCR_dygraph\deploy\hubserving\ocr_rec
方向分类器:.\PaddleOCR_dygraph\deploy\hubserving\ocr_cls
模型库下载地址:https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/doc/doc_ch/models_list.md
* 下载对应的模型后解压,然后修改hubserving目录下,ocr_det\params.py, ocr_rec\params.py, ocr_cls\params.py, ocr_system\params.py四个文件里的模型路径
hub install .\deploy\hubserving\ocr_det\
hub install .\deploy\hubserving\ocr_cls\
hub install .\deploy\hubserving\ocr_rec\
hub install .\deploy\hubserving\ocr_system\
hub serving start -c .\deploy\hubserving\ocr_det\config.json
hub serving start -c .\deploy\hubserving\ocr_cls\config.json
hub serving start -c .\deploy\hubserving\ocr_rec\config.json
hub serving start -c .\deploy\hubserving\ocr_system\config.json
python .\tools\test_hubserving.py http://127.0.0.1:8868/predict/ocr_system .\img.jpg