Merge pull request #10 from PaddlePaddle/develop

mer upstream

PPOCRLabel/README.md

@@ -0,0 +1,80 @@
+# PPOCRLabel
+
+PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，使用python3和pyqt5编写，支持矩形框标注和四点标注模式，导出格式可直接用于PPOCR检测和识别模型的训练。
+
+<img src="./data/gif/steps.gif" width="100%"/>
+
+## 安装
+
+### 1. 安装PaddleOCR
+参考[PaddleOCR安装文档](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md)准备好PaddleOCR
+
+### 2. 安装PPOCRLabel
+#### Windows + Anaconda
+
+下载安装[Anaconda](https://www.anaconda.com/download/#download) (Python 3+)
+
+```
+conda install pyqt=5
+cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
+pyrcc5 -o libs/resources.py resources.qrc
+python PPOCRLabel.py
+```
+
+#### Ubuntu Linux
+
+```
+sudo apt-get install pyqt5-dev-tools
+sudo apt-get install trash-cli
+cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
+sudo pip3 install -r requirements/requirements-linux-python3.txt
+make qt5py3
+python3 PPOCRLabel.py
+```
+
+#### macOS
+```
+pip3 install pyqt5
+pip3 uninstall opencv-python # 由于mac版本的opencv与pyqt有冲突，需先手动卸载opencv
+pip3 install opencv-contrib-python-headless # 安装headless版本的open-cv
+cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
+make qt5py3
+python3 PPOCRLabel.py
+```
+
+## 使用
+
+### 操作步骤
+
+1. 安装与运行：使用上述命令安装与运行程序。
+2. 打开文件夹：在菜单栏点击 “文件” - "打开目录" 选择待标记图片的文件夹<sup>[1]</sup>.
+3. 自动标注：点击 ”自动标注“，使用PPOCR超轻量模型对图片文件名前图片状态<sup>[2]</sup>为 “X” 的图片进行自动标注。
+4. 手动标注：点击 “矩形标注”（推荐直接在英文模式下点击键盘中的 “W”)，用户可对当前图片中模型未检出的部分进行手动绘制标记框。点击键盘P，则使用四点标注模式（或点击“编辑” - “四点标注”），用户依次点击4个点后，双击左键表示标注完成。
+5. 标记框绘制完成后，用户点击 “确认”，检测框会先被预分配一个 “待识别” 标签。
+6. 重新识别：将图片中的所有检测画绘制/调整完成后，点击 “重新识别”，PPOCR模型会对当前图片中的**所有检测框**重新识别<sup>[3]</sup>。
+7. 内容更改：双击识别结果，对不准确的识别结果进行手动更改。
+8. 保存：点击 “保存”，图片状态切换为 “√”，跳转至下一张。
+9. 删除：点击 “删除图像”，图片将会被删除至回收站。
+10. 标注结果：关闭应用程序或切换文件路径后，手动保存过的标签将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “PaddleOCR” - "保存识别结果"后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，识别标签保存在*rec_gt.txt*中<sup>[4]</sup>。
+
+### 注意
+
+[1] PPOCRLabel以文件夹为基本标记单位，打开待标记的图片文件夹后，不会在窗口栏中显示图片，而是在点击 "选择文件夹" 之后直接将文件夹下的图片导入到程序中。
+
+[2] 图片状态表示本张图片用户是否手动保存过，未手动保存过即为 “X”，手动保存过为 “√”。点击 “自动标注”按钮后，PPOCRLabel不会对状态为 “√” 的图片重新标注。
+
+[3] 点击“重新识别”后，模型会对图片中的识别结果进行覆盖。因此如果在此之前手动更改过识别结果，有可能在重新识别后产生变动。
+
+[4] PPOCRLabel产生的文件包括一下几种，请勿手动更改其中内容，否则会引起程序出现异常。
+
+|    文件名     |                             说明                             |
+| :-----------: | :----------------------------------------------------------: |
+|   Label.txt   | 检测标签，可直接用于PPOCR检测模型训练。用户每保存10张检测结果后，程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
+| fileState.txt | 图片状态标记文件，保存当前文件夹下已经被用户手动确认过的图片名称。 |
+|  Cache.cach   |              缓存文件，保存模型自动识别的结果。              |
+|  rec_gt.txt   | 识别标签。可直接用于PPOCR识别模型训练。需用户手动点击菜单栏“PaddleOCR” - "保存识别结果"后产生。 |
+|   crop_img    |   识别数据。按照检测框切割后的图片。与rec_gt.txt同时产生。   |
+
+### 参考资料
+
+1.[Tzutalin. LabelImg. Git code (2015)](https://github.com/tzutalin/labelImg)

PPOCRLabel/build-tools/.gitignore

@@ -0,0 +1,12 @@
+*.spec
+build
+dist
+pyinstaller
+python-2.*
+pywin32*
+virtual-wine
+venv_wine
+PyQt4-*
+lxml-*
+windows_v*
+linux_v*

PPOCRLabel/build-tools/README.md

@@ -0,0 +1,35 @@
+### Deploy to PyPI
+
+```
+cd [ROOT]
+sh build-tools/build-for-pypi.sh
+```
+
+### Build for Ubuntu
+
+```
+cd build-tools
+sh run-in-container.sh
+sh envsetup.sh
+sh build-ubuntu-binary.sh
+```
+
+### Build for Windows
+
+```
+cd build-tools
+sh run-in-container.sh
+sh envsetup.sh
+sh build-windows-binary.sh
+```
+
+### Build for macOS High Sierra
+```
+cd build-tools
+./build-for-macos.sh
+```
+
+Note: If there are some problems, try to
+```
+sudo rm -rf virtual-wne venv_wine
+```

PPOCRLabel/build-tools/build-for-macos.sh

@@ -0,0 +1,30 @@
+#!/bin/sh
+
+brew install python@2
+pip install --upgrade virtualenv
+
+# clone labelimg source
+rm -rf /tmp/labelImgSetup
+mkdir /tmp/labelImgSetup
+cd /tmp/labelImgSetup
+curl https://codeload.github.com/tzutalin/labelImg/zip/master --output labelImg.zip
+unzip labelImg.zip
+rm labelImg.zip
+
+# setup python3 space
+virtualenv --system-site-packages  -p python3 /tmp/labelImgSetup/labelImg-py3
+source /tmp/labelImgSetup/labelImg-py3/bin/activate
+cd labelImg-master
+
+# build labelImg app
+pip install py2app
+pip install PyQt5 lxml
+make qt5py3
+rm -rf build dist
+python setup.py py2app -A
+mv "/tmp/labelImgSetup/labelImg-master/dist/labelImg.app" /Applications
+# deactivate python3
+deactivate
+cd ../
+rm -rf /tmp/labelImgSetup
+echo 'DONE'

PPOCRLabel/build-tools/build-for-pypi.sh

@@ -0,0 +1,17 @@
+#!/bin/sh
+# Packaging and Release
+docker run --workdir=$(pwd)/ --volume="/home/$USER:/home/$USER" tzutalin/py2qt4 /bin/sh -c 'make qt4py2; make test;sudo python setup.py sdist;sudo python setup.py install'
+
+while true; do
+    read -p "Do you wish to deploy this to PyPI(twine upload dist/* or pip install dist/*)?" yn
+    case $yn in
+        [Yy]* ) docker run -it --rm --workdir=$(pwd)/ --volume="/home/$USER:/home/$USER" tzutalin/py2qt4; break;;
+        [Nn]* ) exit;;
+        * ) echo "Please answer yes or no.";;
+    esac
+done
+# python setup.py register
+# python setup.py sdist upload
+# Net pypi: twine upload dist/*
+
+# Test before upladoing: pip install dist/labelImg.tar.gz

PPOCRLabel/build-tools/build-ubuntu-binary.sh

@@ -0,0 +1,24 @@
+#!/bin/bash
+### Ubuntu use pyinstall v3.0
+THIS_SCRIPT_PATH=`readlink -f $0`
+THIS_SCRIPT_DIR=`dirname ${THIS_SCRIPT_PATH}`
+cd pyinstaller
+git checkout v3.2
+cd ${THIS_SCRIPT_DIR}
+
+rm -r build
+rm -r dist
+rm labelImg.spec
+python pyinstaller/pyinstaller.py --hidden-import=xml \
+            --hidden-import=xml.etree \
+            --hidden-import=xml.etree.ElementTree \
+            --hidden-import=lxml.etree \
+             -D -F -n labelImg -c "../labelImg.py" -p ../libs -p ../
+
+FOLDER=$(git describe --abbrev=0 --tags)
+FOLDER="linux_"$FOLDER
+rm -rf "$FOLDER"
+mkdir "$FOLDER"
+cp dist/labelImg $FOLDER
+cp -rf ../data $FOLDER/data
+zip "$FOLDER.zip" -r $FOLDER

PPOCRLabel/build-tools/build-windows-binary.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+### Window requires pyinstall v2.1
+wine msiexec -i python-2.7.8.msi
+wine pywin32-218.win32-py2.7.exe
+wine PyQt4-4.11.4-gpl-Py2.7-Qt4.8.7-x32.exe
+wine lxml-3.7.3.win32-py2.7.exe
+
+THIS_SCRIPT_PATH=`readlink -f $0`
+THIS_SCRIPT_DIR=`dirname ${THIS_SCRIPT_PATH}`
+cd pyinstaller
+git checkout v2.1
+cd ${THIS_SCRIPT_DIR}
+echo ${THIS_SCRIPT_DIR}
+
+#. venv_wine/bin/activate
+rm -r build
+rm -r dist
+rm labelImg.spec
+
+wine c:/Python27/python.exe pyinstaller/pyinstaller.py --hidden-import=xml \
+            --hidden-import=xml.etree \
+            --hidden-import=xml.etree.ElementTree \
+            --hidden-import=lxml.etree \
+             -D -F -n labelImg -c "../labelImg.py" -p ../libs -p ../
+
+FOLDER=$(git describe --abbrev=0 --tags)
+FOLDER="windows_"$FOLDER
+rm -rf "$FOLDER"
+mkdir "$FOLDER"
+cp dist/labelImg.exe $FOLDER
+cp -rf ../data $FOLDER/data
+zip "$FOLDER.zip" -r $FOLDER

PPOCRLabel/build-tools/envsetup.sh

@@ -0,0 +1,53 @@
+#!/bin/sh
+
+THIS_SCRIPT_PATH=`readlink -f $0`
+THIS_SCRIPT_DIR=`dirname ${THIS_SCRIPT_PATH}`
+#OS Ubuntu 14.04
+### Common packages for linux/windows
+if [ ! -e "pyinstaller" ]; then
+    git clone https://github.com/pyinstaller/pyinstaller
+    cd pyinstaller
+    git checkout v2.1 -b v2.1
+    cd ${THIS_SCRIPT_DIR}
+fi
+
+echo "Going to clone and download packages for building windows"
+#Pacakges
+#>  pyinstaller (2.1)
+#>  wine (1.6.2)
+#>  virtual-wine (0.1)
+#>  python-2.7.8.msi
+#>  pywin32-218.win32-py2.7.exe
+
+## tool to install on Ubuntu
+#$ sudo apt-get install wine
+
+### Clone a repo to create virtual wine env
+if [ ! -e "virtual-wine" ]; then
+	git clone https://github.com/htgoebel/virtual-wine.git
+fi
+
+apt-get install scons
+### Create virtual env
+rm -rf venv_wine
+./virtual-wine/vwine-setup venv_wine
+#### Active virutal env
+. venv_wine/bin/activate
+
+### Use wine to install packages to virtual env
+if [ ! -e "python-2.7.8.msi" ]; then
+    wget "https://www.python.org/ftp/python/2.7.8/python-2.7.8.msi"
+fi
+
+if [ ! -e "pywin32-218.win32-py2.7.exe" ]; then
+    wget "http://nchc.dl.sourceforge.net/project/pywin32/pywin32/Build%20218/pywin32-218.win32-py2.7.exe"
+fi
+
+if [ ! -e "PyQt4-4.11.4-gpl-Py2.7-Qt4.8.7-x32.exe" ]; then
+    wget "http://nchc.dl.sourceforge.net/project/pyqt/PyQt4/PyQt-4.11.4/PyQt4-4.11.4-gpl-Py2.7-Qt4.8.7-x32.exe"
+fi
+
+if [ ! -e "lxml-3.7.3.win32-py2.7.exe" ]; then
+    wget "https://pypi.python.org/packages/a3/f6/a28c5cf63873f6c55a3eb7857b736379229b85ba918261d2e88cf886905e/lxml-3.7.3.win32-py2.7.exe#md5=a0f746355876aca4ca5371cb0f1d13ce"
+fi
+

PPOCRLabel/build-tools/run-in-container.sh

@@ -0,0 +1,13 @@
+#!/bin/sh
+docker run -it \
+    --user $(id -u) \
+    -e DISPLAY=unix$DISPLAY \
+    --workdir=$(pwd) \
+    --volume="/home/$USER:/home/$USER" \
+    --volume="/etc/group:/etc/group:ro" \
+    --volume="/etc/passwd:/etc/passwd:ro" \
+    --volume="/etc/shadow:/etc/shadow:ro" \
+    --volume="/etc/sudoers.d:/etc/sudoers.d:ro" \
+    -v /tmp/.X11-unix:/tmp/.X11-unix \
+    tzutalin/py2qt4
+

PPOCRLabel/combobox.py

@@ -0,0 +1,46 @@
+# Copyright (c) <2015-Present> Tzutalin
+# Copyright (C) 2013  MIT, Computer Science and Artificial Intelligence Laboratory. Bryan Russell, Antonio Torralba,
+# William T. Freeman. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and
+# associated documentation files (the "Software"), to deal in the Software without restriction, including without
+# limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
+# Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
+# The above copyright notice and this permission notice shall be included in all copies or substantial portions of
+# the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT
+# NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
+# SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
+# CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+# THE SOFTWARE.
+
+import sys
+try:
+    from PyQt5.QtWidgets import QWidget, QHBoxLayout, QComboBox
+except ImportError:
+    # needed for py3+qt4
+    # Ref:
+    # http://pyqt.sourceforge.net/Docs/PyQt4/incompatible_apis.html
+    # http://stackoverflow.com/questions/21217399/pyqt4-qtcore-qvariant-object-instead-of-a-string
+    if sys.version_info.major >= 3:
+        import sip
+        sip.setapi('QVariant', 2)
+    from PyQt4.QtGui import QWidget, QHBoxLayout, QComboBox
+
+
+class ComboBox(QWidget):
+    def __init__(self, parent=None, items=[]):
+        super(ComboBox, self).__init__(parent)
+
+        layout = QHBoxLayout()
+        self.cb = QComboBox()
+        self.items = items
+        self.cb.addItems(self.items)
+
+        self.cb.currentIndexChanged.connect(parent.comboSelectionChanged)
+
+        layout.addWidget(self.cb)
+        self.setLayout(layout)
+
+    def update_items(self, items):
+        self.items = items
+
+        self.cb.clear()
+        self.cb.addItems(self.items)

PPOCRLabel/libs/__init__.py

@@ -0,0 +1,2 @@
+__version_info__ = ('1', '0', '0')
+__version__ = '.'.join(__version_info__)