PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field. It is written in python3 and pyqt5, supporting rectangular box annotation and four-point annotation modes. Annotations can be directly used for the training of PPOCR detection and recognition models.
PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field, with built-in PPOCR model to automatically detect and re-recognize data. It is written in python3 and pyqt5, supporting rectangular box annotation and four-point annotation modes. Annotations can be directly used for the training of PPOCR detection and recognition models.
<imgsrc="./data/gif/steps_en.gif"width="100%"/>
<imgsrc="./data/gif/steps_en.gif"width="100%"/>
@ -10,11 +10,15 @@ PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field. I
- 2020.12.18: Support re-recognition of a single label box (by [ninetailskim](https://github.com/ninetailskim) ), perfect shortcut keys.
- 2020.12.18: Support re-recognition of a single label box (by [ninetailskim](https://github.com/ninetailskim) ), perfect shortcut keys.
### TODO:
- Lock box mode: For the same scene data, the size and position of the locked detection box can be transferred between different pictures.
- Experience optimization: Add undo, batch operation include move, copy, delete and so on, optimize the annotation process.
## Installation
## Installation
### 1. Install PaddleOCR
### 1. Install PaddleOCR
Refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR
PaddleOCR models has been built in PPOCRLabel, please refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR and make sure it works.
### 2. Install PPOCRLabel
### 2. Install PPOCRLabel
@ -60,7 +64,7 @@ python3 PPOCRLabel.py
4.1 Click 'Create RectBox' or press 'W' in English keyboard mode to draw a new rectangle detection box. Click and release left mouse to select a region to annotate the text area.
4.1 Click 'Create RectBox' or press 'W' in English keyboard mode to draw a new rectangle detection box. Click and release left mouse to select a region to annotate the text area.
4.2 Press 'P' to enter four-point labeling mode which enables you to create any four-point shape by clicking four points with the left mouse button in succession and DOUBLE CLICK the left mouse as the signal of labeling completion.
4.2 Press 'Q' to enter four-point labeling mode which enables you to create any four-point shape by clicking four points with the left mouse button in succession and DOUBLE CLICK the left mouse as the signal of labeling completion.
5. After the marking frame is drawn, the user clicks "OK", and the detection frame will be pre-assigned a "TEMPORARY" label.
5. After the marking frame is drawn, the user clicks "OK", and the detection frame will be pre-assigned a "TEMPORARY" label.
@ -72,7 +76,7 @@ python3 PPOCRLabel.py
9. Click "Delete Image" and the image will be deleted to the recycle bin.
9. Click "Delete Image" and the image will be deleted to the recycle bin.
10. Labeling result: the user can save manually through the menu "File - Save Label", while the program will also save automatically after every 10 images confirmed by the user.the manually checked label will be stored in *Label.txt* under the opened picture folder.
10. Labeling result: the user can save manually through the menu "File - Save Label", while the program will also save automatically after every 5 images confirmed by the user.the manually checked label will be stored in *Label.txt* under the opened picture folder.
Click "PaddleOCR"-"Save Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
Click "PaddleOCR"-"Save Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
### Note
### Note
@ -88,7 +92,7 @@ Therefore, if the recognition result has been manually changed before, it may ch
| Label.txt | The detection label file can be directly used for PPOCR detection model training. After the user saves 10 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
| Label.txt | The detection label file can be directly used for PPOCR detection model training. After the user saves 5 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
| fileState.txt | The picture status file save the image in the current folder that has been manually confirmed by the user. |
| fileState.txt | The picture status file save the image in the current folder that has been manually confirmed by the user. |
| Cache.cach | Cache files to save the results of model recognition. |
| Cache.cach | Cache files to save the results of model recognition. |
| rec_gt.txt | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "File"-"Save recognition result". |
| rec_gt.txt | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "File"-"Save recognition result". |
@ -124,6 +128,15 @@ Therefore, if the recognition result has been manually changed before, it may ch
- Custom model: The model trained by users can be replaced by modifying PPOCRLabel.py in [PaddleOCR class instantiation](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110) referring [Custom Model Code](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/whl_en.md#use-custom-model)
- Custom model: The model trained by users can be replaced by modifying PPOCRLabel.py in [PaddleOCR class instantiation](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110) referring [Custom Model Code](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/whl_en.md#use-custom-model)
### Save
PPOCRLabel supports three ways to save Label.txt
- Automatically save: When it detects that the user has manually checked 5 pictures, the program automatically writes the annotations into Label.txt. The user can change the value of ``self.autoSaveNum`` in ``PPOCRLabel.py`` to set the number of images to be automatically saved after confirmation.
- Manual save: Click "File-Save Marking Results" to manually save the label.
- Close application save
### Export partial recognition results
### Export partial recognition results
For some data that are difficult to recognize, the recognition results will not be exported by **unchecking** the corresponding tags in the recognition results checkbox.
For some data that are difficult to recognize, the recognition results will not be exported by **unchecking** the corresponding tags in the recognition results checkbox.
@ -122,8 +122,7 @@ For a new language request, please refer to [Guideline for new language_requests
<imgsrc="./doc/ppocr_framework.png"width="800">
<imgsrc="./doc/ppocr_framework.png"width="800">
</div>
</div>
PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection, detection frame correction and CRNN text recognition. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module. The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941). Besides, The implementation of the FPGM Pruner and PACT quantization is based on [PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim).
PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection[2], detection frame correction and CRNN text recognition[7]. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module. The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941). Besides, The implementation of the FPGM Pruner [8] and PACT quantization [9] is based on [PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim).
@ -174,7 +173,7 @@ This project is released under <a href="https://github.com/PaddlePaddle/PaddleOC
We welcome all the contributions to PaddleOCR and appreciate for your feedback very much.
We welcome all the contributions to PaddleOCR and appreciate for your feedback very much.
- Many thanks to [Khanh Tran](https://github.com/xxxpsyduck) and [Karl Horky](https://github.com/karlhorky) for contributing and revising the English documentation.
- Many thanks to [Khanh Tran](https://github.com/xxxpsyduck) and [Karl Horky](https://github.com/karlhorky) for contributing and revising the English documentation.
- Many thanks to [zhangxin](https://github.com/ZhangXinNan) for contributing the new visualize function、add .gitgnore and discard set PYTHONPATH manually.
- Many thanks to [zhangxin](https://github.com/ZhangXinNan) for contributing the new visualize function、add .gitignore and discard set PYTHONPATH manually.
- Many thanks to [lyl120117](https://github.com/lyl120117) for contributing the code for printing the network structure.
- Many thanks to [lyl120117](https://github.com/lyl120117) for contributing the code for printing the network structure.
- Thanks [xiangyubo](https://github.com/xiangyubo) for contributing the handwritten Chinese OCR datasets.
- Thanks [xiangyubo](https://github.com/xiangyubo) for contributing the handwritten Chinese OCR datasets.
- Thanks [authorfu](https://github.com/authorfu) for contributing Android demo and [xiadeye](https://github.com/xiadeye) contributing iOS demo, respectively.
- Thanks [authorfu](https://github.com/authorfu) for contributing Android demo and [xiadeye](https://github.com/xiadeye) contributing iOS demo, respectively.
The Style-Text data synthesis tool is a tool based on Baidu's self-developed text editing algorithm "Editing Text in the Wild" [https://arxiv.org/abs/1908.03047](https://arxiv.org/abs/1908.03047).
The Style-Text data synthesis tool is a tool based on Baidu and HUST cooperation research work, "Editing Text in the Wild" [https://arxiv.org/abs/1908.03047](https://arxiv.org/abs/1908.03047).
Different from the commonly used GAN-based data synthesis tools, the main framework of Style-Text includes:
Different from the commonly used GAN-based data synthesis tools, the main framework of Style-Text includes:
* (1) Text foreground style transfer module.
* (1) Text foreground style transfer module.
@ -124,7 +124,7 @@ In actual application scenarios, it is often necessary to synthesize pictures in
* `corpus_file`: Filepath of the corpus. Corpus file should be a text file which will be split by line-endings（'\n'）. Corpus generator samples one line each time.
* `corpus_file`: Filepath of the corpus. Corpus file should be a text file which will be split by line-endings（'\n'）. Corpus generator samples one line each time.
@ -130,10 +130,10 @@ Among them, `paddle` is the Paddle library required for C++ prediction later, an
* After downloading, use the following method to uncompress.
* After downloading, use the following method to uncompress.
```
```
tar -xf fluid_inference.tgz
tar -xf paddle_inference.tgz
```
```
Finally you can see the following files in the folder of `fluid_inference/`.
Finally you can see the following files in the folder of `paddle_inference/`.
## 2. Compile and run the demo
## 2. Compile and run the demo
@ -145,11 +145,11 @@ Finally you can see the following files in the folder of `fluid_inference/`.
```
```
inference/
inference/
|-- det_db
|-- det_db
| |--model
| |--inference.pdparams
| |--params
| |--inference.pdimodel
|-- rec_rcnn
|-- rec_rcnn
| |--model
| |--inference.pdparams
| |--params
| |--inference.pdparams
```
```
@ -188,7 +188,9 @@ cmake .. \
make -j
make -j
```
```
`OPENCV_DIR` is the opencv installation path; `LIB_DIR` is the download (`fluid_inference` folder) or the generated Paddle inference library path (`build/fluid_inference_install_dir` folder); `CUDA_LIB_DIR` is the cuda library file path, in docker; it is `/usr/local/cuda/lib64`; `CUDNN_LIB_DIR` is the cudnn library file path, in docker it is `/usr/lib/x86_64-linux-gnu/`.
`OPENCV_DIR` is the opencv installation path; `LIB_DIR` is the download (`paddle_inference` folder)
or the generated Paddle inference library path (`build/paddle_inference_install_dir` folder);
`CUDA_LIB_DIR` is the cuda library file path, in docker; it is `/usr/local/cuda/lib64`; `CUDNN_LIB_DIR` is the cudnn library file path, in docker it is `/usr/lib/x86_64-linux-gnu/`.
* After the compilation is completed, an executable file named `ocr_system` will be generated in the `build` folder.
* After the compilation is completed, an executable file named `ocr_system` will be generated in the `build` folder.
@ -211,7 +213,6 @@ gpu_id 0 # GPU id when use_gpu is 1
gpu_mem 4000 # GPU memory requested
gpu_mem 4000 # GPU memory requested
cpu_math_library_num_threads 10 # Number of threads when using CPU inference. When machine cores is enough, the large the value, the faster the inference speed
cpu_math_library_num_threads 10 # Number of threads when using CPU inference. When machine cores is enough, the large the value, the faster the inference speed
use_mkldnn 1 # Whether to use mkdlnn library
use_mkldnn 1 # Whether to use mkdlnn library
use_zero_copy_run 1 # Whether to use use_zero_copy_run for inference
max_side_len 960 # Limit the maximum image height and width to 960
max_side_len 960 # Limit the maximum image height and width to 960
det_db_thresh 0.3 # Used to filter the binarized image of DB prediction, setting 0.-0.3 has no obvious effect on the result
det_db_thresh 0.3 # Used to filter the binarized image of DB prediction, setting 0.-0.3 has no obvious effect on the result
@ -244,4 +245,4 @@ The detection results will be shown on the screen, which is as follows.
### 2.3 Notes
### 2.3 Notes
* Paddle2.0.0-beta0 inference model library is recommanded for this tuturial.
* Paddle2.0.0-beta0 inference model library is recommended for this toturial.
weishengyu
4 years ago