基于OpenCv和swing实现图片/视频的展示
图片的展示
swing展示图片，多为操作BufferedImage，这里要关注的核心是将Mat转为BufferedImage。
代码如下：
public Image toBufferedImage(Mat matrix) { int type = BufferedImage.TYPE_BYTE_GRAY; if (matrix.channels() > 1) { type = BufferedImage.TYPE_3BYTE_BGR; } int bufferSize = matrix.channels() * matrix.cols() * matrix.rows(); byte[] buffer = new byte[bufferSize]; matrix.get(0,0,buffer); // get all the pixels BufferedImage image = new BufferedImage(matrix.cols(),matrix.rows(),type); final byte[] targetPixels = ((DataBufferByte)image.getRaster().getDataBuffer()).getData(); System.arraycopy(buffer,0,targetPixels,0,buffer.length); return image; } comment 1：OpenCV Mat --> BufferedImage AWT ， 创建一个byte array用以保存mat 矩阵的像素信息。数组大小为通道数和图片宽/高之积。其中，Mat.get()将所有的元素导入byte数组。最终，图片的光栅信息通过 getDataBuffer()和getDate()组成接收数组，并通过System.arraycopy方法完成填充。实现最终的类型与数据的转移。
comment 2：图片/视频，最终展示的都为BufferedImage，并在JFrame中展示，这里可将toBufferedImage与swing组件配置以展示的部分抽取为类ImageReader。如下：
import org.opencv.core.Mat;import javax.swing.*;import java.awt.*;import java.awt.image.BufferedImage;import java.awt.image.DataBufferByte;/** * @Author: nya * @Description: 图片展示类 * @Date: Created in 10:55 2018/9/21 * @Modify by: */public class ImageViewer { private JLabel imageView; public void show(Mat image,String windowName) { setSystemLookAndFeel(); JFrame frame = createJFrame(windowName); Image loadedImage = toBufferedImage(image); imageView.setIcon(new ImageIcon(loadedImage)); frame.pack(); frame.setLocationRelativeTo(null); frame.setVisible(true); } private JFrame createJFrame(String windowName) { JFrame frame = new JFrame(windowName); imageView = new JLabel(); final JScrollPane imageScrollPane = new JScrollPane(imageView); imageScrollPane.setPreferredSize(new Dimension(640,480)); frame.add(imageScrollPane,BorderLayout.CENTER); frame.setDefaultCloseOperation(WindowConstants.EXIT_ON_CLOSE); return frame; } private void setSystemLookAndFeel() { try { UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName()); } catch (ClassNotFoundException e) { e.printStackTrace(); } catch (InstantiationException e) { e.printStackTrace(); } catch (IllegalAccessException e) { e.printStackTrace(); } catch (UnsupportedLookAndFeelException e) { e.printStackTrace(); } } public Image toBufferedImage(Mat matrix) { int type = BufferedImage.TYPE_BYTE_GRAY; if (matrix.channels() > 1) { type = BufferedImage.TYPE_3BYTE_BGR; } int bufferSize = matrix.channels() * matrix.cols() * matrix.rows(); byte[] buffer = new byte[bufferSize]; matrix.get(0,0,buffer); // get all the pixels BufferedImage image = new BufferedImage(matrix.cols(),matrix.rows(),type); final byte[] targetPixels = ((DataBufferByte)image.getRaster().getDataBuffer()).getData(); System.arraycopy(buffer,0,targetPixels,0,buffer.length); return image; }} image Mat operate code：
String path = "src/main/java/com/opencv/simpleopencvsample/sample/1.jpg"; Mat openFile = null; try { // openFile just is path to mat openFile = openFile(path); Mat clone = openFile.clone(); System.out.println(openFile); Imgproc.resize(openFile,clone,new Size(640,480)); System.out.println(clone); ImageViewer imageViewer = new ImageViewer(); imageViewer.show(clone,"Loaded image"); } catch (Exception e) { e.printStackTrace(); } finally { // comment ： never forget to release the matrix if (openFile != null ) { openFile.release(); } } System.out.println(openFile); 视频的展示
有了图片展示为蓝本，视频的操作关键在于VideoCapture类的使用，展示部分不过是捕获视频的每一帧转为Mat，基于swing循环顺序展示即可。获取mat推荐采用read()，此方法为grab()/retrieve()的结合体。
为适应视频处理的JFrame宽高，此处自定义设置，主要采用toBufferedImage方法。
VideoCapture -> Mat -> display code:
import org.opencv.core.Core;import org.opencv.core.Mat;import org.opencv.videoio.VideoCapture;import org.opencv.videoio.Videoio;import javax.swing.*;import java.awt.*;/** * @Author: nya * @Description: 视频捕获相关操作类VideoCapture使用 * @Date: Created in 13:50 2018/9/21 * @Modify by: */public class VideoCaptureSample { static { System.loadLibrary(Core.NATIVE_LIBRARY_NAME); } private JFrame frame; private JLabel imageLabel; public static void main(String[] args) { VideoCaptureSample sample = new VideoCaptureSample(); sample.initGUI(); sample.runMainLoop(args); } private void initGUI(){ frame = new JFrame("Camera Input Example"); frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE); frame.setSize(555,970); imageLabel = new JLabel(); frame.add(imageLabel); frame.setVisible(true); } private void runMainLoop(String[] args) { ImageViewer viewer = new ImageViewer(); Mat webcamMatImage = new Mat(); Image tempImage; VideoCapture capture = new VideoCapture("src/main/java/com/opencv/simpleopencvsample/sample/1.mp4"); capture.set(Videoio.CAP_PROP_FRAME_WIDTH,550); capture.set(Videoio.CAP_PROP_FRAME_HEIGHT,960); if (capture.isOpened()) { while (true) { capture.read(webcamMatImage); if (!webcamMatImage.empty()) { tempImage = viewer.toBufferedImage(webcamMatImage); ImageIcon imageIcon = new ImageIcon(tempImage,"Captured video"); imageLabel.setIcon(imageIcon); frame.pack(); } else { System.out.println(" --- Frame not captured -- Break !"); break; } } } else { System.out.println("Couldn't open capture."); } }} 常见异常 VideoCapture-isOpened返回false
实际测试中，存在视频路径正常，isOpened()一直返回false的问题。
这是因为视频处理类VideoCapture位于opencv_videoio模块，使用该类时需在运行时加载预先构建的opencv_ffmpeg * .dll / so。该模块如果加载成功，ffmpeg可用于解码/编码视频；否则，使用其它API。在排除路径问题后，仍然无法读取视频则多为此情况。
解决方法很简单，将动态库dll/so导入 管理员/root 权限下配置的java.library.path路径下即可。