mtfyun/rapidocr
1
0 Fork 0
mirror of https://github.com/RapidAI/RapidOCR.git synced 2026-11-10 21:10:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add rapidocr_api module

Browse Source
This commit is contained in:
SWHL committed 2023-05-21 21:20:55 +08:00
1 parent 0c3a528904
commit 05d7e8874e
9 files changed
+505 -123
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

.github/workflows/gen_whl_to_pypi_rapidocr_api.yml
+53
View File
@@ -0,0 +1,53 @@
name: Push rapidocr_api to pypi
on:
push:
branches: [ main ]
paths:
- 'api/rapidocr_api/**'
- 'docs/doc_whl_rapidocr_api.md'
- 'api/setup.py'
- '.github/workflows/gen_whl_to_pypi_rapidocr_api.yml'
jobs:
GenerateWHL_PushPyPi:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Set up Python 3.7
uses: actions/setup-python@v4
with:
python-version: '3.7'
architecture: 'x64'
- name: Set SSH Environment
env:
DEPLOY_KEYS: ${{ secrets.GEN_PYTHON_SDK }}
run: |
mkdir -p ~/.ssh/
echo "$DEPLOY_KEYS" > ~/.ssh/id_rsa
chmod 600 ~/.ssh/id_rsa
chmod 700 ~/.ssh && chmod 600 ~/.ssh/*
- name: Run setup.py
run: |
cd api
pip install -r requirements.txt
python -m pip install --upgrade pip
python setup.py bdist_wheel ${{ github.event.head_commit.message }}
# - name: Publish distribution 📦 to Test PyPI
# uses: pypa/gh-action-pypi-publish@v1.5.0
# with:
# password: ${{ secrets.TEST_PYPI_API_TOKEN }}
# repository_url: https://test.pypi.org/legacy/
# packages_dir: ocrweb/dist/
- name: Publish distribution 📦 to PyPI
uses: pypa/gh-action-pypi-publish@v1.5.0
with:
password: ${{ secrets.RAPIDOCR_API_PYPI }}
packages_dir: api/dist/
api/README.md
+150
View File
@@ -0,0 +1,150 @@
## RapidOCR API
<p>
<a href=""><img src="https://img.shields.io/badge/Python->=3.7,<=3.10-aff.svg"></a>
<a href=""><img src="https://img.shields.io/badge/OS-Linux%2C%20Win%2C%20Mac-pink.svg"></a>
<a href="https://pypi.org/project/rapidocr-api/"><img alt="PyPI" src="https://img.shields.io/pypi/v/rapidocr-api"></a>
<a href="https://pepy.tech/project/rapidocr_api"><img src="https://static.pepy.tech/personalized-badge/rapidocr_api?period=total&units=abbreviation&left_color=grey&right_color=blue&left_text=Downloads"></a>
</p>
- 采用FastAPI+uvicorn实现。
### 使用步骤
1. 安装`rapidocr_api`
```bash
$ pip install rapidocr_api
```
2. 运行
- 用法:
```bash
$ rapidocr_api -h
usage: rapidocr_api [-h] [-ip IP] [-p PORT]
optional arguments:
-h, --help show this help message and exit
-ip IP, --ip IP IP Address
-p PORT, --port PORT IP port
```
- 示例:
```bash
$ rapidocr_api -ip 0.0.0.0 -p 9003
```
3. 使用（本质就是发送一个POST请求，其他语言同理）
- 以文件的方式发送请求
```python
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as f:
file_dict = {'image_file': (img_path, f, 'image/png')}
response = requests.post(url, files=file_dict, timeout=60)
print(response.json())
```
- 以`base64`方式发送post请求
```python
import base64
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as fa:
img_str = base64.b64encode(fa.read())
payload = {'image_data': img_str}
resp = requests.post(url, data=payload)
print(resp.json())
```
4. API输出
- 示例结果：
<details>
<summary>详情</summary>
```json
{
"0": {
"rec_txt": "香港深圳抽血，",
"dt_boxes": [
[265, 18],
[472, 231],
[431, 271],
[223, 59]
],
"score": "0.8175641223788261"
},
"1": {
"rec_txt": "专业查性别",
"dt_boxes": [
[388, 15],
[636, 257],
[587, 307],
[339, 65]
],
"score": "0.8293875356515249"
},
"2": {
"rec_txt": "专业鉴定B超单",
"dt_boxes": [
[215, 84],
[509, 413],
[453, 463],
[159, 134]
],
"score": "0.8626169338822365"
},
"3": {
"rec_txt": "b超仪器查性别",
"dt_boxes": [
[128, 135],
[430, 478],
[366, 534],
[64, 192]
],
"score": "0.8449362441897392"
},
"4": {
"rec_txt": "加微信eee",
"dt_boxes": [
[58, 189],
[268, 450],
[209, 498],
[0, 236]
],
"score": "0.8176911813872201"
},
"5": {
"rec_txt": "可邮寄",
"dt_boxes": [
[493, 261],
[617, 384],
[577, 423],
[454, 300]
],
"score": "0.7494261413812637"
}
}
```
</details>
- 输出结果说明：
- 如果图像中存在文字，则会输出json格式，具体格式介绍如下：
```python
{
"0": {
"rec_txt": "香港深圳抽血，", # 识别的文本
"dt_boxes": [ # 依次为左上角 → 右上角 → 右下角 → 左下角
[265, 18],
[472, 231],
[431, 271],
[223, 59]
],
"score": "0.8175641223788261" # 置信度
}
}
```
- 如果没有检测到文字，则会输出空json(`{}`)。
5. **!!说明：OCR的输出结果为最原始结果，大家可按需进一步扩展。**
api/demo.py
+14
View File
@@ -0,0 +1,14 @@
# -*- encoding: utf-8 -*-
# @Author: SWHL
# @Contact: liekkaskono@163.com
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as f:
file_dict = {'image_file': (img_path, f, 'image/png')}
response = requests.post(url, files=file_dict, timeout=60)
print(response.json())
api/rapidocr_api/__init__.py
+3
View File
@@ -0,0 +1,3 @@
# -*- encoding: utf-8 -*-
# @Author: SWHL
# @Contact: liekkaskono@163.com
api/rapidocr_api/api.py
+81
View File
@@ -0,0 +1,81 @@
# -*- encoding: utf-8 -*-
# @Author: SWHL
# @Contact: liekkaskono@163.com
import argparse
import base64
import io
import json
from pathlib import Path
import sys
import cv2
import numpy as np
import uvicorn
from fastapi import FastAPI, Form, UploadFile
from PIL import Image
from rapidocr_onnxruntime import RapidOCR
sys.path.append(str(Path(__file__).resolve().parent.parent))
class OCRAPIUtils():
def __init__(self) -> None:
self.ocr = RapidOCR()
def __call__(self, img):
img = np.array(img)
img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
ocr_res, _ = self.ocr(img)
if not ocr_res:
return json.dumps({})
out_dict = {str(i): {'rec_txt': rec,
'dt_boxes': dt_box,
'score': score}
for i, (dt_box, rec, score) in enumerate(ocr_res)}
return json.dumps(out_dict, indent=2, ensure_ascii=False)
app = FastAPI()
processor = OCRAPIUtils()
@app.get("/")
async def root():
return {'message': 'Welcome to RapidOCR Server!'}
@app.post('/ocr')
async def ocr(image_file: UploadFile = None, image_data: str = Form(None)):
if image_file:
img = Image.open(image_file.file)
elif image_data:
img_bytes = str.encode(image_data)
img_b64decode = base64.b64decode(img_bytes)
img = Image.open(io.BytesIO(img_b64decode))
else:
raise ValueError(
'When sending a post request, data or files must have a value.')
ocr_res = processor(img)
return ocr_res
def main():
parser = argparse.ArgumentParser('rapidocr_api')
parser.add_argument('-ip', '--ip', type=str, default='0.0.0.0',
help='IP Address')
parser.add_argument('-p', '--port', type=int, default=9003,
help='IP port')
args = parser.parse_args()
cur_file_path = Path(__file__).resolve()
app_path = f'{cur_file_path.parent.name}.{cur_file_path.stem}:app'
print(app_path)
uvicorn.run(app_path, host=args.ip, port=args.port, reload=True)
if __name__ == '__main__':
main()
api/requirements.txt
+6
View File
@@ -0,0 +1,6 @@
requests
rapidocr_onnxruntime
get_pypi_latest_version
wheel
fastapi
uvicorn[standard]
api/setup.py
+67
View File
@@ -0,0 +1,67 @@
# -*- encoding: utf-8 -*-
# @Author: SWHL
# @Contact: liekkaskono@163.com
import sys
from pathlib import Path
import setuptools
from get_pypi_latest_version import GetPyPiLatestVersion
def get_readme():
root_dir = Path(__file__).resolve().parent.parent
readme_path = str(root_dir / 'docs' / 'doc_whl_rapidocr_api.md')
with open(readme_path, 'r', encoding='utf-8') as f:
readme = f.read()
return readme
MODULE_NAME = 'rapidocr_api'
obtainer = GetPyPiLatestVersion()
try:
latest_version = obtainer(MODULE_NAME)
except ValueError:
latest_version = '0.0.1'
VERSION_NUM = obtainer.version_add_one(latest_version)
if len(sys.argv) > 2:
match_str = ' '.join(sys.argv[2:])
matched_versions = obtainer.extract_version(match_str)
if matched_versions:
VERSION_NUM = matched_versions
sys.argv = sys.argv[:2]
setuptools.setup(
name=MODULE_NAME,
version=VERSION_NUM,
platforms='Any',
description='A cross platform OCR API Library based on OnnxRuntime.',
long_description=get_readme(),
long_description_content_type='text/markdown',
author='SWHL',
author_email='liekkaskono@163.com',
url='https://github.com/RapidAI/RapidOCR',
download_url='https://github.com/RapidAI/RapidOCR.git',
license='Apache-2.0',
include_package_data=True,
install_requires=['requests', 'rapidocr_onnxruntime',
'fastapi', 'uvicorn[standard]'],
packages=[MODULE_NAME],
package_data={'': ['*.ico', '*.css', '*.js', '*.html']},
keywords=[
'ocr,text_detection,text_recognition,db,onnxruntime,paddleocr,openvino,rapidocr'
],
classifiers=[
'Programming Language :: Python :: 3.7',
'Programming Language :: Python :: 3.8',
'Programming Language :: Python :: 3.9',
'Programming Language :: Python :: 3.10',
],
python_requires='>=3.7,<=3.10',
entry_points={
'console_scripts': [
f'{MODULE_NAME}={MODULE_NAME}.api:main',
],
},
)
docs/doc_whl_rapidocr_api.md
+130
View File
@@ -0,0 +1,130 @@
## rapidocr-api
<p>
<a href=""><img src="https://img.shields.io/badge/Python->=3.7,<=3.10-aff.svg"></a>
<a href=""><img src="https://img.shields.io/badge/OS-Linux%2C%20Win%2C%20Mac-pink.svg"></a>
<a href="https://pypi.org/project/rapidocr-api/"><img alt="PyPI" src="https://img.shields.io/pypi/v/rapidocr-api"></a>
<a href="https://pepy.tech/project/rapidocr_api"><img src="https://static.pepy.tech/personalized-badge/rapidocr_api?period=total&units=abbreviation&left_color=grey&right_color=blue&left_text=Downloads"></a>
</p>
### Use
1. Install`rapidocr_web[api]`
```bash
$ pip install rapidocr_web[api]
```
2. Run
- Usage:
```bash
$ rapidocr_api -h
usage: rapidocr_api [-h] [-ip IP] [-p PORT]
optional arguments:
-h, --help show this help message and exit
-ip IP, --ip IP IP Address
-p PORT, --port PORT IP port
```
- Example:
```bash
$ rapidocr_api -ip 0.0.0.0 -p 9003
```
3. Use
- Send image data by file format.
```python
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as f:
file_dict = {'image_file': (img_path, f, 'image/png')}
response = requests.post(url, files=file_dict, timeout=60)
print(response.json())
```
- Send image data by base64 format.
```python
import base64
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as fa:
img_str = base64.b64encode(fa.read())
payload = {'image_data': img_str}
resp = requests.post(url, data=payload)
print(resp.json())
```
4. Output
<details>
<summary>Click to expand</summary>
```json
{
"0": {
"rec_txt": "香港深圳抽血，",
"dt_boxes": [
[265, 18],
[472, 231],
[431, 271],
[223, 59]
],
"score": "0.8175641223788261"
},
"1": {
"rec_txt": "专业查性别",
"dt_boxes": [
[388, 15],
[636, 257],
[587, 307],
[339, 65]
],
"score": "0.8293875356515249"
},
"2": {
"rec_txt": "专业鉴定B超单",
"dt_boxes": [
[215, 84],
[509, 413],
[453, 463],
[159, 134]
],
"score": "0.8626169338822365"
},
"3": {
"rec_txt": "b超仪器查性别",
"dt_boxes": [
[128, 135],
[430, 478],
[366, 534],
[64, 192]
],
"score": "0.8449362441897392"
},
"4": {
"rec_txt": "加微信eee",
"dt_boxes": [
[58, 189],
[268, 450],
[209, 498],
[0, 236]
],
"score": "0.8176911813872201"
},
"5": {
"rec_txt": "可邮寄",
"dt_boxes": [
[493, 261],
[617, 384],
[577, 423],
[454, 300]
],
"score": "0.7494261413812637"
}
}
```
</details>
### See details for [RapidOCR](https://github.com/RapidAI/RapidOCR/tree/main/ocrweb).
docs/doc_whl_rapidocr_web.md
+1 -123
View File
@@ -6,8 +6,7 @@
<a href="https://pepy.tech/project/rapidocr_web"><img src="https://static.pepy.tech/personalized-badge/rapidocr_web?period=total&units=abbreviation&left_color=grey&right_color=blue&left_text=Downloads"></a>
</p>
### Web use
### use
1. Install package by pypi.
```bash
$ pip install rapidocr-web
@@ -29,125 +28,4 @@
```
3. Open `http://localhost:9003/` to view, enjoy it.
### API Use
1. Install`rapidocr_web[api]`
```bash
$ pip install rapidocr_web[api]
```
2. Run
- Usage:
```bash
$ rapidocr_api -h
usage: rapidocr_api [-h] [-ip IP] [-p PORT]
optional arguments:
-h, --help show this help message and exit
-ip IP, --ip IP IP Address
-p PORT, --port PORT IP port
```
- Example:
```bash
$ rapidocr_api -ip 0.0.0.0 -p 9003
```
3. Use
- Send image data by file format.
```python
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as f:
file_dict = {'image_file': (img_path, f, 'image/png')}
response = requests.post(url, files=file_dict, timeout=60)
print(response.json())
```
- Send image data by base64 format.
```python
import base64
import requests
url = 'http://localhost:9003/ocr'
img_path = '../python/tests/test_files/ch_en_num.jpg'
with open(img_path, 'rb') as fa:
img_str = base64.b64encode(fa.read())
payload = {'image_data': img_str}
resp = requests.post(url, data=payload)
print(resp.json())
```
4. Output
<details>
<summary>Click to expand</summary>
```json
{
"0": {
"rec_txt": "香港深圳抽血，",
"dt_boxes": [
[265, 18],
[472, 231],
[431, 271],
[223, 59]
],
"score": "0.8175641223788261"
},
"1": {
"rec_txt": "专业查性别",
"dt_boxes": [
[388, 15],
[636, 257],
[587, 307],
[339, 65]
],
"score": "0.8293875356515249"
},
"2": {
"rec_txt": "专业鉴定B超单",
"dt_boxes": [
[215, 84],
[509, 413],
[453, 463],
[159, 134]
],
"score": "0.8626169338822365"
},
"3": {
"rec_txt": "b超仪器查性别",
"dt_boxes": [
[128, 135],
[430, 478],
[366, 534],
[64, 192]
],
"score": "0.8449362441897392"
},
"4": {
"rec_txt": "加微信eee",
"dt_boxes": [
[58, 189],
[268, 450],
[209, 498],
[0, 236]
],
"score": "0.8176911813872201"
},
"5": {
"rec_txt": "可邮寄",
"dt_boxes": [
[493, 261],
[617, 384],
[577, 423],
[454, 300]
],
"score": "0.7494261413812637"
}
}
```
</details>
### See details for [RapidOCR](https://github.com/RapidAI/RapidOCR/tree/main/ocrweb).