README.md 5.22 KB
Newer Older
sugon_cxj's avatar
sugon_cxj committed
1
# paddleocr_paddle_onnxruntime
chenxj's avatar
chenxj committed
2
## 论文
chenxj's avatar
chenxj committed
3
PaddleOCR通过det、rec、cls三个模型分别实现字符检测、字符识别和字符方向分类的应用
chenxj's avatar
chenxj committed
4

chenxj's avatar
chenxj committed
5
6
7
8
9
10
11
12
13
det模型主要用DB算法,参考论文如下:

https://arxiv.org/pdf/1911.08947.pdf

rec模型主要用SVTR算法,参考论文如下:

https://arxiv.org/pdf/2205.00159.pdf

cls模型用mobilenetv3实现通用分类,参考论文如下:
chenxj's avatar
chenxj committed
14

chenxj's avatar
chenxj committed
15
https://arxiv.org/pdf/1905.02244.pdf
sugon_cxj's avatar
sugon_cxj committed
16
## 模型结构
chenxj's avatar
chenxj committed
17
det:
chenxj's avatar
chenxj committed
18

chenxj's avatar
chenxj committed
19
![image](https://developer.hpccube.com/codes/modelzoo/paddleocr/-/raw/main/configs/det/dbnet-arc.png)
chenxj's avatar
chenxj committed
20

chenxj's avatar
chenxj committed
21
rec:
chenxj's avatar
chenxj committed
22

chenxj's avatar
chenxj committed
23
![image](https://developer.hpccube.com/codes/modelzoo/paddleocr/-/raw/main/configs/rec/SVTR-arc.png)
chenxj's avatar
chenxj committed
24

chenxj's avatar
chenxj committed
25
cls:
chenxj's avatar
chenxj committed
26

chenxj's avatar
chenxj committed
27
![image](https://developer.hpccube.com/codes/modelzoo/paddleocr/-/raw/main/configs/cls/mobilenetv3-arc.png)
chenxj's avatar
chenxj committed
28
## 算法原理
sugon_cxj's avatar
sugon_cxj committed
29
![image](https://developer.hpccube.com/codes/modelzoo/paddleocr/-/raw/main/configs/ocr.png)
sugon_cxj's avatar
sugon_cxj committed
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
## 数据集
推荐使用icdar2015数据集[icdar2015](https://rrc.cvc.uab.es/?ch=4&com=downloads)

检测模型训练集文件结构
```
/PaddleOCR/train_data/icdar2015/text_localization/
  └─ icdar_c4_train_imgs/         Training data of icdar dataset
  └─ ch4_test_images/             Testing data of icdar dataset
  └─ train_icdar2015_label.txt    Training annotation of icdar dataset
  └─ test_icdar2015_label.txt     Test annotation of icdar dataset
```
识别模型训练集文件结构
```
|-train_data
  |-rec
    |- rec_gt_train.txt
    |- train
        |- word_001.png
        |- word_002.jpg
        |- word_003.jpg
        | ...
    |-ic15_data
        |- rec_gt_test.txt
        |- test
            |- word_001.jpg
            |- word_002.jpg
            |- word_003.jpg
            | ...
```
chenxj's avatar
chenxj committed
59
## 环境配置
sugon_cxj's avatar
sugon_cxj committed
60
[光源](https://www.sourcefind.cn/#/service-details)可拉取训练以及推理的docker镜像,在[光合开发者社区](https://cancon.hpccube.com:65024/4/main/)可下载paddle、onnxruntime安装包。PaddleOCR推荐的镜像如下:
chenxj's avatar
chenxj committed
61
```
sugon_cxj's avatar
sugon_cxj committed
62
docker pull image.sourcefind.cn:5000/dcu/admin/base/paddlepaddle:2.4.2-centos7.6-dtk-23.04.1-py37-latest
sugon_cxj's avatar
sugon_cxj committed
63
docker run -d -t --privileged --device=/dev/kfd --device=/dev/dri/ --shm-size 64g --network=host --group-add video --name paddleocr-test image.sourcefind.cn:5000/dcu/admin/base/paddlepaddle:2.3.2-centos7.6-dtk-22.10.1-py37-latest
sugon_cxj's avatar
sugon_cxj committed
64
65
66
docker exec -it paddleocr-test bash
pip3 install -r requirements.txt
pip3 install onnxruntime.whl
sugon_cxj's avatar
sugon_cxj committed
67
68
wget -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/pretrained/MobileNetV3_large_x0_5_pretrained.pdparams
wget -P ./pretrain_models/ https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_train.tar
chenxj's avatar
chenxj committed
69
```
chenxj's avatar
chenxj committed
70
## 训练
sugon_cxj's avatar
sugon_cxj committed
71
72

检测模型
chenxj's avatar
chenxj committed
73
```
sugon_cxj's avatar
sugon_cxj committed
74
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/train.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model=./pretrain_models/MobileNetV3_large_x0_5_pretrained
chenxj's avatar
chenxj committed
75
```
sugon_cxj's avatar
sugon_cxj committed
76
识别模型
chenxj's avatar
chenxj committed
77
```
sugon_cxj's avatar
sugon_cxj committed
78
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/train.py -c configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml -o Global.pretrained_model=./pretrain_models/en_PP-OCRv3_rec_train/best_accuracy
chenxj's avatar
chenxj committed
79
```
chenxj's avatar
chenxj committed
80
## 测试
sugon_cxj's avatar
sugon_cxj committed
81
检测模型
chenxj's avatar
chenxj committed
82
```
sugon_cxj's avatar
sugon_cxj committed
83
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/eval.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model=./output/db_mv3/best_accuracy.pdparams
chenxj's avatar
chenxj committed
84
```
sugon_cxj's avatar
sugon_cxj committed
85
识别模型
chenxj's avatar
chenxj committed
86
```
sugon_cxj's avatar
sugon_cxj committed
87
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/eval.py -c configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml -o Global.pretrained_model=./output/v3_en_mobile/best_accuracy.pdparams
chenxj's avatar
chenxj committed
88
```
chenxj's avatar
chenxj committed
89
## 测试(ort)
chenxj's avatar
chenxj committed
90
91
92
93
94
95
96
97
检测模型
```
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/eval.py -c configs/det/det_mv3_db.yml -o Global.pretrained_model=./ch_PP-OCRv3_det_infer/ch_PP-OCRv3_det.onnx --use_onnx=true
```
识别模型
```
python3 -m paddle.distributed.launch --gpus '0,1,2,3'  tools/eval.py -c configs/rec/PP-OCRv3/ch_PP-OCRv3_rec.yml -o Global.pretrained_model=./ch_PP-OCRv3_rec_infer/ch_PP-OCRv3_rec.onnx --use_onnx=true
```
chenxj's avatar
chenxj committed
98
## 推理
chenxj's avatar
chenxj committed
99
```
sugon_cxj's avatar
sugon_cxj committed
100
 python3 tools/infer/predict_system.py --image_dir="./doc/imgs/" --det_model_dir="./ch_PP-OCRv3_det_infer/" --cls_model_dir="./ch_ppocr_mobile_v2.0_cls_infer/" --rec_model_dir="./ch_PP-OCRv3_rec_infer/" --use_angle_cls=true --rec_image_shape=3,48,320 --warmup=1
chenxj's avatar
chenxj committed
101
```
chenxj's avatar
chenxj committed
102
## 推理(ort)
103
104
105
```
python3 tools/infer/predict_system.py --image_dir="./doc/imgs/" --det_model_dir="./ch_PP-OCRv3_det_infer/ch_PP-OCRv3_det.onnx" --cls_model_dir="./ch_ppocr_mobile_v2.0_cls_infer/ch_ppocr_mobile_v2.0_cls_infer.onnx" --rec_model_dir="./ch_PP-OCRv3_rec_infer/ch_PP-OCRv3_rec.onnx" --use_onnx=true --use_angle_cls=true --rec_image_shape=3,48,320 --warmup=1
```
chenxj's avatar
chenxj committed
106
107
## result
![image](https://developer.hpccube.com/codes/modelzoo/paddleocr/-/raw/main/inference_results/08.jpg)
sugon_cxj's avatar
sugon_cxj committed
108
### 精度
sugon_cxj's avatar
sugon_cxj committed
109
110
111
112
113
114
115
116
117
118

检测模型测试
| Model | Precision | Recall |
| :------: | :------: |:------: |
| det | 0.7054 | 0.7193  |

识别模型测试
| Model | Acc | 
| :------: | :------: |
| rec | 0.6490 | 
chenxj's avatar
chenxj committed
119
120
121
122
123
124
125
126
127
128

检测模型测试(ort)
| Model | Precision | Recall |
| :------: | :------: |:------: |
| det | 0.5097 | 0.4068  |

识别模型测试(ort)
| Model | Acc | 
| :------: | :------: |
| rec | 0.6076 | 
chenxj's avatar
chenxj committed
129
130
131
132
## 应用场景
### 算法类别
ocr
### 热点应用行业
sugon_cxj's avatar
sugon_cxj committed
133
工业制造,金融,交通,教育,医疗
chenxj's avatar
chenxj committed
134
## 源码仓库及问题反馈
chenxj's avatar
chenxj committed
135
https://developer.hpccube.com/codes/modelzoo/paddleocr
sugon_cxj's avatar
sugon_cxj committed
136
## 参考资料
sugon_cxj's avatar
sugon_cxj committed
137
138
* [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)