安卓中使用OpenCV的Houghlines

Question

我正在尝试使用opencv3对四边形物体进行透视校正。我成功地显示了线条，并使用Imgproc.HoughLinesP()实现了Houghlines，并尝试使用Imgproc.lines()突出显示线条，但输出没有成功。以下是我的代码以及我附加的输出图像。

请告诉我出了什么问题，应该怎么做...

Mat initImg; // initial image
Mat greyImg; // converted to grey

Mat lines = new Mat();
int threshold = 50;
int minLineSize = 20;
int lineGap = 10;

initImg = Imgcodecs.imread(imgLoc, 1);
greyImg = new Mat();
Imgproc.cvtColor(initImg, greyImg, Imgproc.COLOR_BGR2GRAY);
Bitmap bitm = Bitmap.createBitmap(greyImg.cols(), greyImg.rows(),Bitmap.Config.ARGB_8888);

Imgproc.blur(greyImg, greyImg, new Size(3.d, 3.d));
Imgproc.adaptiveThreshold(greyImg, greyImg, 255, Imgproc.ADAPTIVE_THRESH_MEAN_C, Imgproc.THRESH_BINARY_INV, 15, 4);

Imgproc.HoughLinesP(greyImg, lines, 1, Math.PI/180, threshold,
        minLineSize, lineGap);
// lines returns rows x columns and rows is always 1. I dont know why please help me to understand
for (int x = 0; x < lines.cols(); x++) {

    double[] vec = lines.get(0, x);
    double[] val = new double[4];

    double x1 = vec[0],
            y1 = vec[1],
            x2 = vec[2],
            y2 = vec[3];

    System.out.println(TAG+"Coordinates: x1=>"+x1+" y1=>"+y1+" x2=>"+x2+" y2=>"+y2);
    Point start = new Point(x1, y1);
    Point end = new Point(x2, y2);

    Imgproc.line(greyImg, start, end, new Scalar(0,255, 0, 255), 3);

}

Utils.matToBitmap(greyImg, bitm);

if(bitm!=null){
    Toast.makeText(getApplicationContext(), "Bitmap not null", Toast.LENGTH_SHORT).show();
    iv.setImageBitmap(bitm);
}else{
    Toast.makeText(getApplicationContext(), "Bitmap null", Toast.LENGTH_SHORT).show();
}

我的输出：

- Robz

Answer 1

我终于成功获取了霍夫线。在灰度图像中，霍夫线没有显示出来，但我在灰度图像中提取出了霍夫线，然后在彩色图像上绘制直线，这样就成功了。

Imgproc.HoughLinesP(greyImg, lines, 1, Math.PI/180, threshold,
                minLineSize, lineGap);

        for (int x = 0; x < lines.rows(); x++)
        {
            double[] vec = lines.get(x, 0);
            double x1 = vec[0],
                    y1 = vec[1],
                    x2 = vec[2],
                    y2 = vec[3];
            Point start = new Point(x1, y1);
            Point end = new Point(x2, y2);
            double dx = x1 - x2;
            double dy = y1 - y2;

            double dist = Math.sqrt (dx*dx + dy*dy);

            if(dist>300.d)  // show those lines that have length greater than 300
                Imgproc.line(initImg, start, end, new Scalar(0,255, 0, 255),5);// here initimg is the original image.

        }

- Robz