Opencv's Canny Edge Detection in C++

OpenCV's Canny Edge Detection in C++

Change this line

cvtColor( image, gray_image, CV_RGB2GRAY );

to

std::vector<cv::Mat> channels;
cv::Mat hsv;
cv::cvtColor( image, hsv, CV_RGB2HSV );
cv::split(hsv, channels);
gray_image = channels[0];

The problem seems to be that your hand in gray scale is very close to the gray background. I have applied Canny on the hue (color) because the skin color should be sufficiently different.

Also, the Canny thresholds look a bit crazy. The accepted norm is that the higher one should be 2x to 3x the lower. 350 is a bit too much and it doesn't help solve the main problem.

Edit

with these thresholds I was able to extract quite a good contour

cv::Canny(image,contours,35,90);

Reading a bit of theory about the algorithm will help you understand what happens and what you should do to improve. wiki canny on google

However, the improvement above will give you much better results (provided you use better thresholds than 10, 350. Try (40, 120) )

How to process the result of OpenCV:: Canny Edge Detection

edges is array (Mat) of the same size as source picture, and it contains zero pixels and max value (255) pixels at edges that Canny function found

You can emphasize edges at source image (by white color), making bitwise_or operation with edges (in my). Or transform edges to color image to draw color edges. Here is example of using edges as mask.

Edges is raster result. To get set of segments, use findContours function on edges, then you can use drawContours

Note that this information is well-googlable.

Get angle from OpenCV Canny edge detector

canny doesn't give you this directly.
However, you can calculate the angle from the Sobel transform, which is used internally in canny().

Pseudo code:

    cv::Canny(image,contours,10,350);
    cv::Sobel(image, dx, CV_64F, 1, 0, 3, 1, 0, cv::BORDER_REPLICATE);
    cv::Sobel(image, dy, CV_64F, 0, 1, 3, 1, 0, cv::BORDER_REPLICATE);

    cv::Mat angle(image.size(), CV_64F)

    foreach (i,j) such that contours[i, j] > 0
    {
        angle[i, j] = atan2(dy[i,j], dx[i , j])
    }

OpenCV's Canny Detection slow on more complicated image

That behavior is expected, indeed.

There are 3 ways to speed things up:

• OpenCV can benefit from an Intel processor, if you have one you can install Intel IPP. You might have to compile OpenCV on your own to enable this feature.

• Use OpenCV's GPU module. The method gpu::Canny() provides an implementation of Canny that runs on the GPU. OpenCV can run certain algorithms on the GPU if your video card supports CUDA or OpenCL. You might have to compile OpenCV on your own to enable this feature.

• Investigate a different approach: sometimes a different set of algorithms can achieve the same result in a smaller amount of time. I talked briefly about this on your other question.

---

Related Topics