Hi Fika,
Based on my understanding, you want to detect the edges of an image and extract them without any noise. You can go through the following steps and the attached code snippets to achieve the desired results:
1. Read and preprocess the image by converting it into grayscale using ‘rgb2gray’ function.
% Read the image
image = imread('img.jpeg');
grayImage = rgb2gray(image);
2. Apply a gaussian filter using ‘imgaussfilt’ function, which is essential for smoothing the image for effective edge detection. Compute the gradients in the x and y directions using 'sobel’ operator .
% Apply Gaussian filter
blurredImage = imgaussfilt(grayImage, 1.4);
[Gx, Gy] = imgradientxy(blurredImage, 'sobel');
magnitude = sqrt(Gx.^2 + Gy.^2);
direction = atan2(Gy, Gx);
3. Suppress the non-maximum pixels for thinning the edges by comparing each pixel’s gradient magnitude with its neighbours.
% Non-Maximum Suppression
nms = zeros(size(magnitude));
[rows, cols] = size(magnitude);
for i = 2:rows-1
for j = 2:cols-1
angle = direction(i, j) * 180 / pi;
angle = mod(angle + 180, 180);
if ((angle >= 0 && angle < 22.5) || (angle >= 157.5 && angle <= 180))
if (magnitude(i, j) >= magnitude(i, j-1) && magnitude(i, j) >= magnitude(i, j+1))
nms(i, j) = magnitude(i, j);
end
elseif (angle >= 22.5 && angle < 67.5)
if (magnitude(i, j) >= magnitude(i-1, j+1) && magnitude(i, j) >= magnitude(i+1, j-1))
nms(i, j) = magnitude(i, j);
end
elseif (angle >= 67.5 && angle < 112.5)
if (magnitude(i, j) >= magnitude(i-1, j) && magnitude(i, j) >= magnitude(i+1, j))
nms(i, j) = magnitude(i, j);
end
elseif (angle >= 112.5 && angle < 157.5)
if (magnitude(i, j) >= magnitude(i-1, j-1) && magnitude(i, j) >= magnitude(i+1, j+1))
nms(i, j) = magnitude(i, j);
end
end
end
end
4. Apply double thresholding to classify pixels as strong edges, weak edges, or non-edges. This helps in distinguishing between significant edges and noise.
highThreshold = max(nms(:)) * 0.1;
lowThreshold = highThreshold * 0.5;
strongEdges = nms >= highThreshold;
weakEdges = (nms >= lowThreshold) & (nms < highThreshold);
5. Track the edges by starting from strong edges then to connected weak edges to preserve important edges while reducing noise. Then display the results.
% Edge tracking by hysteresis
edges = zeros(size(nms));
for i = 2:rows-1
for j = 2:cols-1
if strongEdges(i, j)
edges(i, j) = 1;
elseif weakEdges(i, j)
if any(any(strongEdges(i-1:i+1, j-1:j+1)))
edges(i, j) = 1;
end
end
end
end
You can refer to the output below for a better understanding:
For more information, kindly refer to the documentation links below:
- ‘rgb2gray’: https://www.mathworks.com/help/matlab/ref/rgb2gray.html
- ‘imgaussfilt’: https://www.mathworks.com/help/images/ref/imgaussfilt.html
- ‘imgradientxy’: https://www.mathworks.com/help/images/ref/imgradientxy.html
I hope this works for you!
