java平台下如何使用opencv计算HSV直方图？


how to use opencv to calculate HSV histogram in java platform?

实际上，我想找出图像中的主色，所以我想找到图像的 HSV 直方图，从而过滤掉其他颜色。但是，我不知道如何在使用 opence 的 java 平台中执行此操作。我只在 C 中找到代码。谢谢。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
    Mat image = Highgui.imread("binary07.jpg");
    //Mat src = new Mat(image.height(), image.width(), CvType.CV_8UC2);

    Imgproc.cvtColor(image, image, Imgproc.COLOR_RGB2GRAY);

    List<Mat> hsv_planes = new ArrayList<Mat>();
    Core.split(image, hsv_planes);

    MatOfInt histSize = new MatOfInt(256);

    final MatOfFloat histRange = new MatOfFloat(0f, 256f);

    boolean accumulate = false;

    Mat h_hist = new Mat();
    Mat s_hist = new Mat();
    Mat v_hist = new Mat();

    //error appear in the following sentences

    Imgproc.calcHist((List<Mat>) hsv_planes.get(0), new MatOfInt(3), new Mat(), h_hist, histSize, histRange, accumulate);
    Imgproc.calcHist((List<Mat>) hsv_planes.get(1), new MatOfInt(3), new Mat(), s_hist, histSize, histRange, accumulate);
    Imgproc.calcHist((List<Mat>) hsv_planes.get(2), new MatOfInt(3), new Mat(), v_hist, histSize, histRange, accumulate);

    int hist_w = 512;
    int hist_h = 600;
    long bin_w = Math.round((double) hist_w / 256);
    //bin_w = Math.round((double) (hist_w / 256));

    Mat histImage = new Mat(hist_h, hist_w, CvType.CV_8UC1);
    Core.normalize(h_hist, h_hist, 3, histImage.rows(), Core.NORM_MINMAX);
    Core.normalize(s_hist, s_hist, 3, histImage.rows(), Core.NORM_MINMAX);
    Core.normalize(v_hist, v_hist, 3, histImage.rows(), Core.NORM_MINMAX);


    for (int i = 1; i < 256; i++) {
        Point p1 = new Point(bin_w * (i - 1), hist_h - Math.round(h_hist.get(i - 1, 0)[0]));
        Point p2 = new Point(bin_w * (i), hist_h - Math.round(h_hist.get(i, 0)[0]));
        Core.line(histImage, p1, p2, new Scalar(255, 0, 0), 2, 8, 0);

        Point p3 = new Point(bin_w * (i - 1), hist_h - Math.round(s_hist.get(i - 1, 0)[0]));
        Point p4 = new Point(bin_w * (i), hist_h - Math.round(s_hist.get(i, 0)[0]));
        Core.line(histImage, p3, p4, new Scalar(0, 255, 0), 2, 8, 0);

        Point p5 = new Point(bin_w * (i - 1), hist_h - Math.round(v_hist.get(i - 1, 0)[0]));
        Point p6 = new Point(bin_w * (i), hist_h - Math.round(v_hist.get(i, 0)[0]));
        Core.line(histImage, p5, p6, new Scalar(0, 0, 255), 2, 8, 0);

    }

    Highgui.imwrite("histogram.jpg", histImage);

我不知道如何得到拆分函数后的输出。

参考：
http://docs.opencv.org/java/
http://docs.opencv.org/doc/tutorials/imgproc/histograms/histogram_calculation/histogram_calculation.html

在代码中，颜色转换：

1
Imgproc.cvtColor(image, image, Imgproc.COLOR_RGB2GRAY);

应该是HSV而不是灰色：

1
Imgproc.cvtColor(image, image, Imgproc.COLOR_BGR2HSV);

在您的示例中，您将只有一个(灰色)平面，而不是 3 个 HSV 通道。当您访问第 2 和第 3 平面时，这会出错。

这里是将源图像的直方图与 OpenCV 2.4.11 Java (Android) 的参考图像进行比较的代码。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
      // Assume SourceImage is a Bitmap ARGB_8888
      BitmapFactory.Options options = new BitmapFactory.Options();
      options.inPreferredConfig = Bitmap.Config.ARGB_8888;
      Bitmap refImage = BitmapFactory.decodeFile(mBaseDir +"some_reference.jpg", options);

      Mat hsvRef = new Mat();
      Mat hsvSource = new Mat();

      Mat srcRef = new Mat(refImage.getHeight(), refImage.getWidth(), CvType.CV_8U, new Scalar(4));
      Utils.bitmapToMat(refImage, srcRef);


      Mat srcSource = new Mat(SourceImage.getHeight(), SourceImage.getWidth(), CvType.CV_8U, new Scalar(4));
      Utils.bitmapToMat(SourceImage, srcSource);

      /// Convert to HSV
      Imgproc.cvtColor(srcRef, hsvRef, Imgproc.COLOR_BGR2HSV);
      Imgproc.cvtColor(srcSource, hsvSource, Imgproc.COLOR_BGR2HSV);

      /// Using 50 bins for hue and 60 for saturation
      int hBins = 50;
      int sBins = 60;
      MatOfInt histSize = new MatOfInt( hBins,  sBins);

      // hue varies from 0 to 179, saturation from 0 to 255
      MatOfFloat ranges =  new MatOfFloat( 0f,180f,0f,256f );

      // we compute the histogram from the 0-th and 1-st channels
      MatOfInt channels = new MatOfInt(0, 1);



      Mat histRef = new Mat();
      Mat histSource = new Mat();

      ArrayList<Mat> histImages=new ArrayList<Mat>();
      histImages.add(hsvRef);
      Imgproc.calcHist(histImages,
              channels,
              new Mat(),
              histRef,
              histSize,
              ranges,
              false);
      Core.normalize(histRef,
              histRef,
              0,
              1,
              Core.NORM_MINMAX,
              -1,
              new Mat());

      histImages=new ArrayList<Mat>();
      histImages.add(hsvSource);
      Imgproc.calcHist(histImages,
              channels,
              new Mat(),
              histSource,
              histSize,
              ranges,
              false);
      Core.normalize(histSource,
              histSource,
              0,
              1,
              Core.NORM_MINMAX,
              -1,
              new Mat());

      double resp1 = Imgproc.compareHist(histRef, histSource, 0);
      double resp2 = Imgproc.compareHist(histRef, histSource, 1);
      double resp3 = Imgproc.compareHist(histRef, histSource, 2);
      double resp4 = Imgproc.compareHist(histRef, histSource, 3);

下一个代码适用于一个深度通道。您只需进行一些修改即可添加其他两个频道

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
//Calculate histogram
java.util.List<Mat> matList = new LinkedList<Mat>();
matList.add(imageIR_gray);
Mat histogram = new Mat();
MatOfFloat ranges=new MatOfFloat(0,256);
MatOfInt histSize = new MatOfInt(255);
Imgproc.calcHist(
                matList,
                new MatOfInt(0),
                new Mat(),
                histogram ,
                histSize ,
                ranges);

// Create space for histogram image
Mat histImage = Mat.zeros( 100, (int)histSize.get(0, 0)[0], CvType.CV_8UC1);
// Normalize histogram                          
Core.normalize(histogram, histogram, 1, histImage.rows() , Core.NORM_MINMAX, -1, new Mat() );  
// Draw lines for histogram points
for( int i = 0; i < (int)histSize.get(0, 0)[0]; i++ )
{                  
        Core.line(
                histImage,
                new org.opencv.core.Point( i, histImage.rows() ),
                new org.opencv.core.Point( i, histImage.rows()-Math.round( histogram.get(i,0)[0] )) ,
                new Scalar( 255, 255, 255),
                1, 8, 0 );
}