2022-04-18 15:28:22 +08:00
|
|
|
|
# PaddleOCR 快速开始
|
2021-09-06 23:59:41 +08:00
|
|
|
|
|
2022-04-19 18:09:38 +08:00
|
|
|
|
**说明:** 本文主要介绍PaddleOCR wheel包对PP-OCR系列模型的快速使用,如要体验文档分析相关功能,请参考[PP-Structure快速使用教程](../../ppstructure/docs/quickstart.md)。
|
2022-04-18 16:22:26 +08:00
|
|
|
|
|
2022-03-30 11:32:22 +08:00
|
|
|
|
- [1. 安装](#1)
|
|
|
|
|
- [1.1 安装PaddlePaddle](#11)
|
|
|
|
|
- [1.2 安装PaddleOCR whl包](#12)
|
2022-02-28 10:45:58 +08:00
|
|
|
|
- [2. 便捷使用](#2)
|
|
|
|
|
- [2.1 命令行使用](#21)
|
2021-09-03 21:32:00 +08:00
|
|
|
|
- [2.1.1 中英文模型](#211)
|
|
|
|
|
- [2.1.2 多语言模型](#212)
|
2022-02-28 10:45:58 +08:00
|
|
|
|
- [2.2 Python脚本使用](#22)
|
2021-09-03 21:32:00 +08:00
|
|
|
|
- [2.2.1 中英文与多语言使用](#221)
|
2022-03-30 11:32:22 +08:00
|
|
|
|
- [3.小结](#3)
|
2021-09-03 21:48:06 +08:00
|
|
|
|
|
2020-10-13 17:49:16 +08:00
|
|
|
|
|
2021-09-03 21:32:00 +08:00
|
|
|
|
<a name="1"></a>
|
2022-03-30 11:32:22 +08:00
|
|
|
|
## 1. 安装
|
|
|
|
|
|
|
|
|
|
<a name="11"></a>
|
|
|
|
|
### 1.1 安装PaddlePaddle
|
|
|
|
|
|
|
|
|
|
> 如果您没有基础的Python运行环境,请参考[运行环境准备](./environment.md)。
|
|
|
|
|
|
|
|
|
|
- 您的机器安装的是CUDA9或CUDA10,请运行以下命令安装
|
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
- 您的机器是CPU,请运行以下命令安装
|
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
更多的版本需求,请参照[飞桨官网安装文档](https://www.paddlepaddle.org.cn/install/quick)中的说明进行操作。
|
|
|
|
|
|
|
|
|
|
<a name="12"></a>
|
|
|
|
|
### 1.2 安装PaddleOCR whl包
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
pip install "paddleocr>=2.0.1" # 推荐使用2.0.1+版本
|
2020-07-10 23:22:32 +08:00
|
|
|
|
```
|
|
|
|
|
|
2022-04-18 15:28:22 +08:00
|
|
|
|
- 对于Windows环境用户:直接通过pip安装的shapely库可能出现`[winRrror 126] 找不到指定模块的问题`。建议从[这里](https://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely)下载shapely安装包完成安装。
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="2"></a>
|
2021-09-03 19:20:07 +08:00
|
|
|
|
## 2. 便捷使用
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="21"></a>
|
2021-09-03 19:20:07 +08:00
|
|
|
|
### 2.1 命令行使用
|
|
|
|
|
|
2021-09-06 11:36:24 +08:00
|
|
|
|
PaddleOCR提供了一系列测试图片,点击[这里](https://paddleocr.bj.bcebos.com/dygraph_v2.1/ppocr_img.zip)下载并解压,然后在终端中切换到相应目录
|
2020-07-10 23:22:32 +08:00
|
|
|
|
|
|
|
|
|
```
|
2021-09-03 19:20:07 +08:00
|
|
|
|
cd /path/to/ppocr_img
|
2020-07-10 23:22:32 +08:00
|
|
|
|
```
|
|
|
|
|
|
2022-04-18 16:22:26 +08:00
|
|
|
|
如果不使用提供的测试图片,可以将下方`--image_dir`参数替换为相应的测试图片路径。
|
|
|
|
|
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="211"></a>
|
2021-09-03 19:20:07 +08:00
|
|
|
|
#### 2.1.1 中英文模型
|
|
|
|
|
|
2021-11-09 20:01:05 +08:00
|
|
|
|
* 检测+方向分类器+识别全流程:`--use_angle_cls true`设置使用方向分类器识别180度旋转文字,`--use_gpu false`设置不使用GPU
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
```bash
|
2022-05-07 21:36:57 +08:00
|
|
|
|
paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true --use_gpu false
|
2021-09-03 19:20:07 +08:00
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
结果是一个list,每个item包含了文本框,文字和识别置信度
|
|
|
|
|
|
|
|
|
|
```bash
|
2022-04-29 11:38:59 +08:00
|
|
|
|
[[[28.0, 37.0], [302.0, 39.0], [302.0, 72.0], [27.0, 70.0]], ('纯臻营养护发素', 0.9658738374710083)]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
......
|
|
|
|
|
```
|
|
|
|
|
|
2022-10-08 16:28:30 +08:00
|
|
|
|
此外,paddleocr也支持输入pdf文件,并且可以通过指定参数`page_num`来控制推理前面几页,默认为0,表示推理所有页。
|
|
|
|
|
```bash
|
|
|
|
|
paddleocr --image_dir ./xxx.pdf --use_angle_cls true --use_gpu false --page_num 2
|
|
|
|
|
```
|
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
- 单独使用检测:设置`--rec`为`false`
|
|
|
|
|
|
|
|
|
|
```bash
|
|
|
|
|
paddleocr --image_dir ./imgs/11.jpg --rec false
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
结果是一个list,每个item只包含文本框
|
|
|
|
|
|
|
|
|
|
```bash
|
2022-04-29 11:38:59 +08:00
|
|
|
|
[[27.0, 459.0], [136.0, 459.0], [136.0, 479.0], [27.0, 479.0]]
|
|
|
|
|
[[28.0, 429.0], [372.0, 429.0], [372.0, 445.0], [28.0, 445.0]]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
......
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
- 单独使用识别:设置`--det`为`false`
|
|
|
|
|
|
|
|
|
|
```bash
|
2022-05-07 21:36:57 +08:00
|
|
|
|
paddleocr --image_dir ./imgs_words/ch/word_1.jpg --det false
|
2021-09-03 19:20:07 +08:00
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
结果是一个list,每个item只包含识别结果和识别置信度
|
|
|
|
|
|
|
|
|
|
```bash
|
2022-04-29 11:38:59 +08:00
|
|
|
|
['韩国小馆', 0.994467]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
```
|
|
|
|
|
|
2022-05-30 16:10:37 +08:00
|
|
|
|
**版本说明**
|
2023-08-06 11:17:13 +08:00
|
|
|
|
paddleocr默认使用PP-OCRv4模型(`--ocr_version PP-OCRv4`),如需使用其他版本可通过设置参数`--ocr_version`,具体版本说明如下:
|
2022-05-30 16:10:37 +08:00
|
|
|
|
| 版本名称 | 版本说明 |
|
|
|
|
|
| --- | --- |
|
2023-08-06 11:17:13 +08:00
|
|
|
|
| PP-OCRv4 | 支持中、英文检测和识别,方向分类器,支持多语种识别 |
|
2022-05-30 16:10:37 +08:00
|
|
|
|
| PP-OCRv3 | 支持中、英文检测和识别,方向分类器,支持多语种识别 |
|
2022-05-30 17:25:21 +08:00
|
|
|
|
| PP-OCRv2 | 支持中英文的检测和识别,方向分类器,多语言暂未更新 |
|
2022-05-30 16:10:37 +08:00
|
|
|
|
| PP-OCR | 支持中、英文检测和识别,方向分类器,支持多语种识别 |
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
2022-05-30 16:10:37 +08:00
|
|
|
|
如需新增自己训练的模型,可以在[paddleocr](../../paddleocr.py)中增加模型链接和字段,重新编译即可。
|
|
|
|
|
|
|
|
|
|
更多whl包使用可参考[whl包文档](./whl.md)
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="212"></a>
|
2021-09-23 20:58:14 +08:00
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
#### 2.1.2 多语言模型
|
|
|
|
|
|
2022-05-07 21:36:57 +08:00
|
|
|
|
PaddleOCR目前支持80个语种,可以通过修改`--lang`参数进行切换,对于英文模型,指定`--lang=en`。
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
``` bash
|
2022-05-07 21:36:57 +08:00
|
|
|
|
paddleocr --image_dir ./imgs_en/254.jpg --lang=en
|
2020-07-10 23:22:32 +08:00
|
|
|
|
```
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
<div align="center">
|
|
|
|
|
<img src="../imgs_en/254.jpg" width="300" height="600">
|
|
|
|
|
<img src="../imgs_results/multi_lang/img_02.jpg" width="600" height="600">
|
|
|
|
|
</div>
|
|
|
|
|
|
|
|
|
|
结果是一个list,每个item包含了文本框,文字和识别置信度
|
|
|
|
|
|
|
|
|
|
```text
|
2022-04-29 11:38:59 +08:00
|
|
|
|
[[[67.0, 51.0], [327.0, 46.0], [327.0, 74.0], [68.0, 80.0]], ('PHOCAPITAL', 0.9944712519645691)]
|
|
|
|
|
[[[72.0, 92.0], [453.0, 84.0], [454.0, 114.0], [73.0, 122.0]], ('107 State Street', 0.9744491577148438)]
|
|
|
|
|
[[[69.0, 135.0], [501.0, 125.0], [501.0, 156.0], [70.0, 165.0]], ('Montpelier Vermont', 0.9357033967971802)]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
......
|
2020-07-10 23:22:32 +08:00
|
|
|
|
```
|
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
常用的多语言简写包括
|
2020-07-10 23:22:32 +08:00
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
| 语种 | 缩写 | | 语种 | 缩写 | | 语种 | 缩写 |
|
|
|
|
|
| -------- | ----------- | ---- | -------- | ------ | ---- | -------- | ------ |
|
|
|
|
|
| 中文 | ch | | 法文 | fr | | 日文 | japan |
|
|
|
|
|
| 英文 | en | | 德文 | german | | 韩文 | korean |
|
|
|
|
|
| 繁体中文 | chinese_cht | | 意大利文 | it | | 俄罗斯文 | ru |
|
2020-07-10 23:22:32 +08:00
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
全部语种及其对应的缩写列表可查看[多语言模型教程](./multi_languages.md)
|
2021-09-07 10:01:06 +08:00
|
|
|
|
|
2021-09-06 11:08:07 +08:00
|
|
|
|
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="22"></a>
|
2021-09-03 19:20:07 +08:00
|
|
|
|
### 2.2 Python脚本使用
|
2021-09-03 21:48:06 +08:00
|
|
|
|
<a name="221"></a>
|
2021-09-03 19:20:07 +08:00
|
|
|
|
#### 2.2.1 中英文与多语言使用
|
|
|
|
|
|
2021-09-06 23:59:41 +08:00
|
|
|
|
通过Python脚本使用PaddleOCR whl包,whl包会自动下载ppocr轻量级模型作为默认模型。
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
* 检测+方向分类器+识别全流程
|
|
|
|
|
|
|
|
|
|
```python
|
|
|
|
|
from paddleocr import PaddleOCR, draw_ocr
|
|
|
|
|
|
|
|
|
|
# Paddleocr目前支持的多语言语种可以通过修改lang参数进行切换
|
|
|
|
|
# 例如`ch`, `en`, `fr`, `german`, `korean`, `japan`
|
|
|
|
|
ocr = PaddleOCR(use_angle_cls=True, lang="ch") # need to run only once to download and load model into memory
|
|
|
|
|
img_path = './imgs/11.jpg'
|
|
|
|
|
result = ocr.ocr(img_path, cls=True)
|
2022-10-08 16:28:30 +08:00
|
|
|
|
for idx in range(len(result)):
|
|
|
|
|
res = result[idx]
|
|
|
|
|
for line in res:
|
|
|
|
|
print(line)
|
2021-09-03 19:20:07 +08:00
|
|
|
|
|
|
|
|
|
# 显示结果
|
|
|
|
|
from PIL import Image
|
2022-10-08 16:28:30 +08:00
|
|
|
|
result = result[0]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
image = Image.open(img_path).convert('RGB')
|
|
|
|
|
boxes = [line[0] for line in result]
|
|
|
|
|
txts = [line[1][0] for line in result]
|
|
|
|
|
scores = [line[1][1] for line in result]
|
|
|
|
|
im_show = draw_ocr(image, boxes, txts, scores, font_path='./fonts/simfang.ttf')
|
|
|
|
|
im_show = Image.fromarray(im_show)
|
|
|
|
|
im_show.save('result.jpg')
|
|
|
|
|
```
|
|
|
|
|
|
|
|
|
|
结果是一个list,每个item包含了文本框,文字和识别置信度
|
2020-10-13 17:49:16 +08:00
|
|
|
|
|
|
|
|
|
```bash
|
2022-04-29 11:38:59 +08:00
|
|
|
|
[[[28.0, 37.0], [302.0, 39.0], [302.0, 72.0], [27.0, 70.0]], ('纯臻营养护发素', 0.9658738374710083)]
|
2021-09-03 19:20:07 +08:00
|
|
|
|
......
|
2020-07-10 23:22:32 +08:00
|
|
|
|
```
|
|
|
|
|
|
2021-09-03 19:20:07 +08:00
|
|
|
|
结果可视化
|
|
|
|
|
|
|
|
|
|
<div align="center">
|
|
|
|
|
<img src="../imgs_results/whl/11_det_rec.jpg" width="800">
|
|
|
|
|
</div>
|
2021-09-07 10:01:06 +08:00
|
|
|
|
|
2022-03-30 11:32:22 +08:00
|
|
|
|
|
|
|
|
|
<a name="3"></a>
|
|
|
|
|
|
2022-10-08 16:28:30 +08:00
|
|
|
|
如果输入是PDF文件,那么可以参考下面代码进行可视化
|
|
|
|
|
|
|
|
|
|
```python
|
|
|
|
|
from paddleocr import PaddleOCR, draw_ocr
|
|
|
|
|
|
|
|
|
|
# Paddleocr目前支持的多语言语种可以通过修改lang参数进行切换
|
|
|
|
|
# 例如`ch`, `en`, `fr`, `german`, `korean`, `japan`
|
2023-02-23 16:31:07 +08:00
|
|
|
|
ocr = PaddleOCR(use_angle_cls=True, lang="ch", page_num=2) # need to run only once to download and load model into memory
|
2022-10-08 16:28:30 +08:00
|
|
|
|
img_path = './xxx.pdf'
|
|
|
|
|
result = ocr.ocr(img_path, cls=True)
|
|
|
|
|
for idx in range(len(result)):
|
|
|
|
|
res = result[idx]
|
|
|
|
|
for line in res:
|
|
|
|
|
print(line)
|
|
|
|
|
|
|
|
|
|
# 显示结果
|
|
|
|
|
import fitz
|
|
|
|
|
from PIL import Image
|
|
|
|
|
import cv2
|
|
|
|
|
import numpy as np
|
|
|
|
|
imgs = []
|
|
|
|
|
with fitz.open(img_path) as pdf:
|
|
|
|
|
for pg in range(0, pdf.pageCount):
|
|
|
|
|
page = pdf[pg]
|
|
|
|
|
mat = fitz.Matrix(2, 2)
|
|
|
|
|
pm = page.getPixmap(matrix=mat, alpha=False)
|
|
|
|
|
# if width or height > 2000 pixels, don't enlarge the image
|
|
|
|
|
if pm.width > 2000 or pm.height > 2000:
|
|
|
|
|
pm = page.getPixmap(matrix=fitz.Matrix(1, 1), alpha=False)
|
|
|
|
|
|
|
|
|
|
img = Image.frombytes("RGB", [pm.width, pm.height], pm.samples)
|
|
|
|
|
img = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR)
|
|
|
|
|
imgs.append(img)
|
|
|
|
|
for idx in range(len(result)):
|
|
|
|
|
res = result[idx]
|
|
|
|
|
image = imgs[idx]
|
|
|
|
|
boxes = [line[0] for line in res]
|
|
|
|
|
txts = [line[1][0] for line in res]
|
|
|
|
|
scores = [line[1][1] for line in res]
|
|
|
|
|
im_show = draw_ocr(image, boxes, txts, scores, font_path='doc/fonts/simfang.ttf')
|
|
|
|
|
im_show = Image.fromarray(im_show)
|
|
|
|
|
im_show.save('result_page_{}.jpg'.format(idx))
|
|
|
|
|
```
|
|
|
|
|
|
2022-03-30 11:32:22 +08:00
|
|
|
|
## 3. 小结
|
|
|
|
|
|
|
|
|
|
通过本节内容,相信您已经熟练掌握PaddleOCR whl包的使用方法并获得了初步效果。
|
|
|
|
|
|
2023-08-06 11:17:13 +08:00
|
|
|
|
飞桨AI套件(PaddleX)提供了飞桨生态优质模型,是训压推一站式全流程高效率开发平台,其使命是助力AI技术快速落地,愿景是使人人成为AI Developer!目前PP-OCRv4已上线PaddleX,您可以进入[通用OCR](https://aistudio.baidu.com/aistudio/modelsdetail?modelId=286)体验模型训练、压缩和推理部署全流程。
|