OpenCV Android Create New Image Using the Edges of the Largest Contour

Question

I am able to detect the largest square/rectangle (in green) in an image. However, I want to convert the largest square/rectangle detected in the image into a new image (to be stored in a new Mat).

Here's the return image of this function that has the largest rectangle/square on it: http://img153.imageshack.us/img153/9308/nn4w.png

Here is my code so far:

private Mat findLargestRectangle(Mat original_image) {
    Mat imgSource = original_image;

    //convert the image to black and white
    Imgproc.cvtColor(imgSource, imgSource, Imgproc.COLOR_BGR2GRAY);

    //convert the image to black and white does (8 bit)
    Imgproc.Canny(imgSource, imgSource, 50, 50);

    //apply gaussian blur to smoothen lines of dots
    Imgproc.GaussianBlur(imgSource, imgSource, new Size(5, 5), 5);

    //find the contours
    List<MatOfPoint> contours = new ArrayList<MatOfPoint>();
    Imgproc.findContours(imgSource, contours, new Mat(), Imgproc.RETR_LIST, Imgproc.CHAIN_APPROX_SIMPLE);

    double maxArea = -1;
    int maxAreaIdx = -1;
    MatOfPoint temp_contour = contours.get(0); //the largest is at the index 0 for starting point
    MatOfPoint2f approxCurve = new MatOfPoint2f();
    Mat largest_contour = contours.get(0);
    List<MatOfPoint> largest_contours = new ArrayList<MatOfPoint>();
    for (int idx = 0; idx < contours.size(); idx++) {
        temp_contour = contours.get(idx);
        double contourarea = Imgproc.contourArea(temp_contour);
        //compare this contour to the previous largest contour found
        if (contourarea > maxArea) {
            //check if this contour is a square
            MatOfPoint2f new_mat = new MatOfPoint2f( temp_contour.toArray() );
            int contourSize = (int)temp_contour.total();
            Imgproc.approxPolyDP(new_mat, approxCurve, contourSize*0.05, true);
            if (approxCurve.total() == 4) {
                maxArea = contourarea;
                maxAreaIdx = idx;
                largest_contours.add(temp_contour);
                largest_contour = temp_contour;
            }
        }
    }
    MatOfPoint temp_largest = largest_contours.get(largest_contours.size()-1);
    largest_contours = new ArrayList<MatOfPoint>();
    largest_contours.add(temp_largest);

    Imgproc.cvtColor(imgSource, imgSource, Imgproc.COLOR_BayerBG2RGB);
    Imgproc.drawContours(imgSource, largest_contours, -1, new Scalar(0, 255, 0), 1);

    //create the new image here using the largest detected square

    Toast.makeText(getApplicationContext(), "Largest Contour: ", Toast.LENGTH_LONG).show();

    return imgSource;
}

The variable largest_contours is a list of MatOfPoint but only contains the largest contour, also stored in largest_contour variable. How can I possible make a new image from the largest contour?

I am using OpenCV in Android and there are few tutorials for detecting images but not exactly on how to use Imgproc.warpPerspective()

Thanks!

Answer 1

All you need is to find the corners of this contour. You can use extreme points approach.

You should simply find out the point having minimum x and minimum y (this is your topleft), minimum x maximum y (this is your bottom left), and so on.

In C++, there is a library named algoritm which has min/max methods. For instance min_element will help you to find the point having minimum x or y. Don't forget to include header.

After having your 4 points, you can use perspective transform.

First input your points to this method. The destination should be

Point2f dest[4] = {(0,0),(image.width,0),(0,image.height),(image.height,image.width)}

for your case. The matrix (M) you obtain from this method will transform your points to destination using another method.

Good luck.

Edit: On the second thought; since your contour is not a parallelogram, you have your extreme points as min x = bottomleft, min y = topleft, and so on. Then it is easier to find the minimum element.

Answer 2

That's not the way of using drawContour on a new Mat. This document will help you.

Correct way would be

    Imgproc.findContours(imgSource, contours, new Mat(), Imgproc.RETR_LIST, Imgproc.CHAIN_APPROX_SIMPLE);

    ...

    Imgproc.cvtColor(imgSource, imgSource, Imgproc.COLOR_BayerBG2RGB);
    Imgproc.drawContours(imgSource, contours, maxAreaIdx, new Scalar(0, 255, 0), 1);