Simple example of using OpenCV in a Vaadin application.
This application takes a picture using web browsers camera API (available in modern browsers) and runs OpenCV face recognition algorithm (using CascadeClassifier ) for it. If a face is detected a "troll face" is added on top of it.
This application was inspired by the ingenious "Trollator" mobile Android application.
You can find a blog post about installing OpenCV for Java and Maven in my Dzone article.
- OpenCV Installation for local Maven repository
OpenCV is a native library with Java bindings so you need to install this to your system.
- libopencv_java300.so installed in you java.library.path (
- opencv-300.jar availble for application
There are good instructions how to build OpenCV with Java bindings for your own platform here: http://docs.opencv.org/doc/tutorials/introduction/desktop_java/java_dev_intro.html
Once you have built the Java library you can install the resulting jar file to your local Maven repository using
mvn install:install-file -Dfile=./bin/opencv-300.jar
-DgroupId=org.opencv -DartifactId=opencv -Dversion=3.0.0 -Dpackaging=jar
- Building this application
Once OpenCV jar library is available as a local Maven dependency, you can clone and build this application simply using Git and Maven:
git clone https://github.com/samie/trollator.git
mvn install
And run the application using the embedded Jetty plugin in http://localhost:8888
mvn jetty:run