[Solved] Computer-Vision project2-Edge detection

To practise edge detections

1. Write edge detection routines for Sobel Edge, Prewitt Edge and Canny Edge operators.

• Apply Sobel Kernel and for Sobel Edges with a

reasonable threshold.

• Apply Prewitt Kernel and for Prewitt Edges with a

resonable threshold.

• Canny Edge Detector requires longer multiple steps discussed in class.
  1. Create a 1D Gaussian mask G to convolve with your image (I). The standard deviations(s) of this Gaussian is a parameter to the edge detector (call it >0).
  2. Create a 1D mask for the first derivative of the Gaussian in the x and y directions; call these G_x and G_y. The same >0 value is used as in step a)
  3. Convolve your image I with G along the rows to give the x component image (I_x) and down the columns to give the y component image (I_y).
  4. Convolve I_x with Gx to give I_x, the x component of I convolve with the derivative of the Gaussian, and convolve I_y with G_y to give I_y, y component of I convolved with the derivative of the Gaussian.
  5. Compute the magnitude of the edge response by combining the x and y components. The magnitude of the result can be computed at each pixel (x, y) as: .
  6. Implement non-maximum suppression algorithm (the center pixel must have a larger gradient magnitude than its neighbors in the gradient direction) that we discussed in the lecture. Pixles that are not local maxima should

be removed with this method. In other words, not all the pixels indicating strong magnitude are edges in fact. We need to remove false-positive edge locations from the image.

1. Finally apply Hysteresis thresholding to obtain the final edge-map.

2. You need to generate intermediate images for each step to check how your algorithm works.
3. Try images uploaded to class page and your favorite images. For more images, refer

https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/bsds/BSDS300/html/datas et/images.html. You may calculate precision and recall for Berkeley data.

4. Report your experiments with analytic document.

Discussions:

1. Analyze result images from three different implementation
2. What are effects of to images?
3. Does Canny Edge Detector achieve the three goals?