I’ve been looking into edge detection in Python and it seems that the simplest way to achieve this is to have an image with blacks and whites as the main components. After reading through some websites with some examples it seems that although it does seem possible to apply it to color images, it seems to get much more complicated rather quickly. I found a piece of code that scans an image and detects pixel values and one could make it so that the program would go through the image from top to bottom thus mapping the location of each point that represents a boundary from black to white and use that to make an outline of said edge. This shape could be stored as a vector and then the vector could be used to calculate the angle in the detected edge region (tying back into my angle of repose project).

From: http://lu-vision-dock.blogspot.com/2009/04/sobel-edge-detection.html

def sobel(img):
if img.mode!= "RGB":
      img = img.convert("RGB")
  out_image = Image.new(img.mode, img.size, None)
  imgdata = img.load()
  outdata = out_image.load()

  gx = [[-1, 0, 1], [-2, 0, 2], [-1, 0, 1]]
  gy = [[-1, -2, -1], [0, 0, 0], [1, 2, 1]]

for row inxrange(1, img.size[0]-1):
for col inxrange(1, img.size[1]-1):
          pixel_gx = pixel_gy = 0
          pxval = sum(imgdata[row,col])/3
for i inrange(-1, 2):
for j inrange(-1, 2):
                  val = sum(imgdata[row+i,col+j])/3
                  pixel_gx += gx[i+1][j+1]* val
                  pixel_gy += gy[i+1][j+1]* val
          newpixel = math.sqrt(pixel_gx * pixel_gx + pixel_gy * pixel_gy)
          newpixel = 255 - int(newpixel)
          outdata[row, col] = (newpixel, newpixel, newpixel)
return out_image

This is still pretty new to me and I’ve yet to wrap my head around that the math is actually doing but I think vectors are a good way to go with this one. Should be fun.

Aside from this there are a lot of actually dedicated image processing modules out there which I’ve yet to look into thoroughly. It’s likely that one of them contains something that will simplify what I’m trying to achieve so I’ll get to looking into those as well.

A couple of promising prospects are OpenCV and Scikits-image (adds image processing functionality to SciPy). Scikits-image has some particular examples which are very interesting and can probably be put to use in what I’m trying to achieve.

http://scikits-image.org/docs/dev/auto_examples/applications/plot_coins_segmentation.html

http://scikits-image.org/docs/dev/auto_examples/plot_contours.html

http://scikits-image.org/docs/dev/auto_examples/plot_canny.html

http://scikits-image.org/docs/dev/auto_examples/plot_watershed.html