OpenCV in advanced Android development: Edge detection

In the last Android development tutorial we’ve cover light source detection using OpenCV. This time we’ll cover edge detection, which allows us to distinguish objects from background. This task is usually the starting point to more advanced content analysis (e.g. determining object features, quantities etc.).

Edge detection OpenCV

Again, OpenCV makes it easier by providing the Imgproc.Canny() method, which does all the work for us. The method itself, nomen omen, uses Canny’s algorithm to perform this task.

private void detectEdges(Bitmap bitmap) {
	Mat rgba = new Mat();
	Utils.bitmapToMat(bitmap, rgba);

	Mat edges = new Mat(rgba.size(), CvType.CV_8UC1);
	Imgproc.cvtColor(rgba, edges, Imgproc.COLOR_RGB2GRAY, 4);
	Imgproc.Canny(edges, edges, 80, 100);

	// Don't do that at home or work it's for visualization purpose.
	BitmapHelper.showBitmap(this, bitmap, imageView);
	Bitmap resultBitmap = Bitmap.createBitmap(edges.cols(), edges.rows(), Bitmap.Config.ARGB_8888);
	Utils.matToBitmap(edges, resultBitmap);
	BitmapHelper.showBitmap(this, resultBitmap, detectEdgesImageView);
}

The process is pretty straight-forward:

1. Read the bitmap
2. Covert to greyscale
3. Apply Canny’s algorithm

As always, we start by create an array from the initial bitmap.

Mat rgba = new Mat();
Utils.bitmapToMat(bitmap, rgba);

Next make an array, in which all the calculation results will be saved, and convert the picture to greyscale. And at the end use the Imgproc.Canny() method to detect edges.

Mat edges = new Mat(rgba.size(), CvType.CV_8UC1);
Imgproc.cvtColor(rgba, edges, Imgproc.COLOR_RGB2GRAY, 4);
Imgproc.Canny(edges, edges, 80, 100);

The Imgproc.Canny() method takes the following parameters:

• Mat image – 8-bit input image
• Mat edges – output edge map; single channels 8-bit image, which has the same size as image
• double treshold1– first threshold for the hysteresis procedure
• dobule treshold2– second threshold for the hysteresis procedure

Now we can display two pictures: the original one and the other that is the result of our Imgproc.Canny() method.

// Don't do that at home or work it's for visualization purpose.
BitmapHelper.showBitmap(this, bitmap, imageView);
Bitmap resultBitmap = Bitmap.createBitmap(edges.cols(), edges.rows(), Bitmap.Config.ARGB_8888);
Utils.matToBitmap(edges, resultBitmap);
BitmapHelper.showBitmap(this, resultBitmap, detectEdgesImageView);

And that is it!

OpenCV: Edge detection

As you can see, OpenCV can deliver great effects with minimum effort when building your mobile app. It’s a great and useful library, which provides loads of solutions out-of-the-box.

A code sample that we’ve used in a commercial Android application can be found on our GitHub page.