Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Facial Recognition Using Java

DZone's Guide to

Facial Recognition Using Java

Learn how to use the Sarxos library and the Openimaj library in order to perform facial recognition on images from a webcam.

· AI Zone
Free Resource

Find out how AI-Fueled APIs from Neura can make interesting products more exciting and engaging. 

In this post, we will learn how to extract faces out of an image from a webcam. We will make use of two libraries: Sarxos and Openimaj.

  • Language used: Java

  • Program usedFaceDetector.java  

  • Git repo: Link

  • Website: Link

Here's our POM dependency: 

<dependency>
	<groupId>org.openimaj</groupId>
	<artifactId>image-feature-extraction</artifactId>
	<version>1.3.5</version>
</dependency>
<dependency>
	<artifactId>faces</artifactId>
	<groupId>org.openimaj</groupId>
	<version>1.3.5</version>
	<scope>compile</scope>
</dependency>
<dependency>
	<groupId>com.github.sarxos</groupId>
	<artifactId>webcam-capture</artifactId>
	<version>0.3.11</version>
	<scope>test</scope>
</dependency>  

These are our variables:

      private static final long serialVersionUID = 1L;  
      private static final HaarCascadeDetector detector = new HaarCascadeDetector();  
      private Webcam webcam = null;  
      private BufferedImage img= null;  
      private List<DetectedFace> faces = null;  

Main method:

 public static void main(String[] args) throws IOException {
  new FaceDetector().detectFace();
 }

Using the Sarxos Library

We create an object with the FaceDetector class, which classes the default constructor. Then, we call the detectFace method of this class.

FaceDetector constructor:

      public FaceDetector() throws IOException {
       webcam = Webcam.getDefault();
       webcam.setViewSize(WebcamResolution.VGA.getSize());
       webcam.open(true);
       img = webcam.getImage();
       webcam.close();
       ImagePanel panel = new ImagePanel(img);
       panel.setPreferredSize(WebcamResolution.VGA.getSize());
       add(panel);
       setTitle("Face Recognizer");
       setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
       pack();
       setLocationRelativeTo(null);
       setVisible(true);
      }

How it works:

  1. Use the Sarxos library for webcam here.

  2. Create a webcam object and set the viewsize.

  3. Open the webcam using the open method.

  4. Take the image from the webcam and store it in a BufferedImage object named img.

  5. Close the webcam and pass the image obtained in the ImagePanel class, which will then be added to Frame.

  6. Show the frame to the user with the webcam image that will be processed.

Using the Openimaj Library

detectFace method:

      public void detectFace() {
       JFrame fr = new JFrame("Discovered Faces");
       faces = detector.detectFaces(ImageUtilities.createFImage(img));
       if (faces == null) {
        System.out.println("No faces found in the captured image");
        return;
       }
       Iterator < DetectedFace > dfi = faces.iterator();
       while (dfi.hasNext()) {
        DetectedFace face = dfi.next();
        FImage image1 = face.getFacePatch();
        ImagePanel p = new ImagePanel(ImageUtilities.createBufferedImage(image1));
        fr.add(p);
       }
       fr.setLayout(new FlowLayout(0));
       fr.setSize(500, 500);
       fr.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
       fr.setVisible(true);
      }

How it works:

  1. Use the Ppenimaj library for face detection.

  2. Create a new Frame that will show the results.

  3. Use the detectFaces method of the HaarCascadeDetector class object detector, passing the image to be processed. ImageUtilities is used to create FImage out of BufferedImage.

  4. If no face is found in the image, an error message is returned.

  5. Otherwise, iterate through each face and retrieve the faces using the getFacePatch method.

  6. Use the createBufferedImage method of ImageUtilities class again — this time, to get a BufferedImage out of FImage.

  7. Add all the faces to the resulting frame.

ImagePanel Class:

 package com.cooltrickshome;
 import java.awt.Dimension;
 import java.awt.Graphics;
 import java.awt.Image;
 import javax.swing.ImageIcon;
 import javax.swing.JPanel;
 class ImagePanel
 extends JPanel {
  private Image img;
  public ImagePanel(String img) {
   this(new ImageIcon(img).getImage());
  }
  public ImagePanel(Image img) {
   this.img = img;
   Dimension size = new Dimension(img.getWidth(null), img.getHeight(null));
   setPreferredSize(size);
   setMinimumSize(size);
   setMaximumSize(size);
   setSize(size);
   setLayout(null);
  }
  public void paintComponent(Graphics g) {
   g.drawImage(this.img, 0, 0, null);
  }
 }

How it works:

  1. This is used to show the image on a panel.

Output:

Full Program

FaceDetector.java:

 package com.cooltrickshome;
 /**  
  * Reference:  
  * https://github.com/sarxos/webcam-capture/tree/master/webcam-capture-examples/webcam-capture-detect-face  
  * http://openimaj.org/  
  */
 import java.awt.FlowLayout;
 import java.awt.image.BufferedImage;
 import java.io.IOException;
 import java.util.Iterator;
 import java.util.List;
 import javax.swing.JFrame;
 import org.openimaj.image.FImage;
 import org.openimaj.image.ImageUtilities;
 import org.openimaj.image.processing.face.detection.DetectedFace;
 import org.openimaj.image.processing.face.detection.HaarCascadeDetector;
 import com.github.sarxos.webcam.Webcam;
 import com.github.sarxos.webcam.WebcamResolution;
 public class FaceDetector extends JFrame {
  private static final long serialVersionUID = 1 L;
  private static final HaarCascadeDetector detector = new HaarCascadeDetector();
  private Webcam webcam = null;
  private BufferedImage img = null;
  private List < DetectedFace > faces = null;
  public FaceDetector() throws IOException {
   webcam = Webcam.getDefault();
   webcam.setViewSize(WebcamResolution.VGA.getSize());
   webcam.open(true);
   img = webcam.getImage();
   webcam.close();
   ImagePanel panel = new ImagePanel(img);
   panel.setPreferredSize(WebcamResolution.VGA.getSize());
   add(panel);
   setTitle("Face Recognizer");
   setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
   pack();
   setLocationRelativeTo(null);
   setVisible(true);
  }
  public void detectFace() {
   JFrame fr = new JFrame("Discovered Faces");
   faces = detector.detectFaces(ImageUtilities.createFImage(img));
   if (faces == null) {
    System.out.println("No faces found in the captured image");
    return;
   }
   Iterator < DetectedFace > dfi = faces.iterator();
   while (dfi.hasNext()) {
    DetectedFace face = dfi.next();
    FImage image1 = face.getFacePatch();
    ImagePanel p = new ImagePanel(ImageUtilities.createBufferedImage(image1));
    fr.add(p);
   }
   fr.setLayout(new FlowLayout(0));
   fr.setSize(500, 500);
   fr.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
   fr.setVisible(true);
  }
  public static void main(String[] args) throws IOException {
   new FaceDetector().detectFace();
  }
 }

ImagePanel.java:

 package com.cooltrickshome;
 import java.awt.Dimension;
 import java.awt.Graphics;
 import java.awt.Image;
 import javax.swing.ImageIcon;
 import javax.swing.JPanel;
 class ImagePanel
 extends JPanel {
  private Image img;
  public ImagePanel(String img) {
   this(new ImageIcon(img).getImage());
  }
  public ImagePanel(Image img) {
   this.img = img;
   Dimension size = new Dimension(img.getWidth(null), img.getHeight(null));
   setPreferredSize(size);
   setMinimumSize(size);
   setMaximumSize(size);
   setSize(size);
   setLayout(null);
  }
  public void paintComponent(Graphics g) {
   g.drawImage(this.img, 0, 0, null);
  }
 }

I hope this helps!

References

To find out how AI-Fueled APIs can increase engagement and retention, download Six Ways to Boost Engagement for Your IoT Device or App with AI today.

Topics:
ai ,facial recognition ,tutorial ,java

Published at DZone with permission of Anurag Jain, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}