Hello, OpenCV and Java

Checking if you have Java

Open the terminal (ctrl + alt + T).

Run the following command to check if you already have Java:

$ java -version

You are good to go if you see something like this:

java version "1.8.0_181"
Java(TM) SE Runtime Environment (build 1.8.0_181-b13)
Java HotSpot(TM) 64-Bit Server VM (build 25.181-b13, mixed mode)

You may have another version of Oracle's Java or OpenJDK, but this should not be a problem.

If you do not see the message above or something similar, you will need to install Java. If you can see the message, jump to the next section.

Run the following commands to install Oracle's Java 8:

$ sudo add-apt-repository ppa:webupd8team/java
$ sudo apt-get update
$ sudo apt-get install oracle-java8-installer

Run the java -version command again to make sure everything is fine.

Download OpenCV

Run the following command to install the OpenCV library:

$ sudo apt-get install libopencv-dev

That's it.

Creating Our First Program

Using your favorite text editor, create a file named HelloOpenCVJava.java. I am using Vim, so the command would be:

$ vim HelloOpenCVJava.java

Now, copy and paste the following code:

import org.opencv.core.Core;

public class HelloOpenCVJava {

  /** Loading the OpenCV library. */
  static {
    System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
  }

  public static void main(String[] args) {

    /** Printing the OpenCV version. */
    System.out.println("Hello OpenCV " + Core.VERSION + "!");

    /** Printing the Java version. */
    System.out.println("Hello to you too, Java " + System.getProperty("java.version") + "!");
  }

}

Let's breakdown the HelloOpenCVJava application.

The import keyword is necessary to load the Core class from the OpenCV library.

We use the NATIVE_LIBRARY_NAME property of the Core class to find the corresponding name of the library and use as parameter to the loadLibrary method. In this case, it returns opencv_java320. This will allow us to use the OpenCV library. After that, we print the versions of the library and Java.

Compiling and Running HelloOpenCVJava

Your project directory must be looking like this right now:

.
|__ HelloOpenCVJava.java

If it is, run the following command to compile HelloOpenCVJava.java:

$ javac -cp /usr/share/OpenCV/java/opencv-320.jar HelloOpenCVJava.java

Now you should have your bytecode class file HelloOpenCVJava.class. Here's how the directory should look like:

.
|__ HelloOpenCVJava.class
|__ HelloOpenCVJava.java

The -cp flag tells the compiler to set the system property java.class.path, which is a list of directories, JAR files and ZIP files that contain class files. We are using here to indicate that we want to find our opencv-320.jar.

We have everything we need to run our application now. We do this with the following command:

$ java -Djava.library.path=/usr/lib/jni -cp /usr/share/OpenCV/java/opencv-320.jar:. HelloOpenCVJava

-Djava.library.path sets the java.library.path property (-D is a flag to set a system property) to inform the JVM where it can locate native libraries. This property is part of the system environment used by Java, in order to find and load native libraries used by an application.

This should be your output:

Hello OpenCV 3.2.0!
Hello to you too, Java 1.8.0_181!

The second line may change depending on your version of Java.

We did it! Well, I did it. I hope that you can replicate this in your machine.

To finish, I would like to point out that in the OpenCV Java Tutorials Documentation there's a pretty neat tutorial on how to use OpenCV in an IDE like Eclipse, if you prefer to work in such fashion.

In the next chapter of this journey, we will play a little with Mat objects and some images. Stay tuned!