Image2Letters

from an image by haar cascade get plate, then from plate get regions (by MSER) , from regions get letters/digits by SVM

Usage:

Haar cascade description:

https://archive.ics.uci.edu/ml/datasets/Letter+Recognition

• opencv installation examples are in Kauppi:~/Programs/opencv-3.2.0/build/bin

see http://docs.opencv.org/3.1.0/d7/d9f/tutorial_linux_install.html opencv installation (because cv2.text needs opencv>= 3.0) I downloaded 06.06.2017 https://github.com/opencv/opencv/archive/3.2.0.zip https://github.com/opencv/opencv_contrib/archive/3.2.0.zip

sudo apt-get install apt-file apt-file update apt-file search gstreamer-base-1.0 sudo apt-get install libgstreamer1.0-dev sudo apt-get install libgstreamer-plugins-base1.0-dev sudo apt-get install libavresample-dev libavresample-ffmpeg2 libgphoto2-dev sudo apt-get install libgoogle-glog-dev sudo apt-get install libopenblas-dev liblapacke-dev checkinstall

Kauppi:~/Programs/opencv-3.2.0 cd /Programs/opencv-3.2.0 mkdir build cd build cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local
-D OPENCV_EXTRA_MODULES_PATH=/Programs/opencv_contrib-3.2.0/modules
-D INSTALL_PYTHON_EXAMPLES=ON -D ENABLE_PRECOMPILED_HEADERS=OFF
-D BUILD_EXAMPLES=ON ..

make

sudo make install sudo ldconfig

for tesseract (not used for the moment)

you need tesserocr (and tesseract, which hopefully comes automatically below))

apt-get install tesseract-ocr libtesseract-dev libleptonica-dev sudo -H pip3 install tesserocr