how to add color to certain areas of a .png image?

Hi all, i have been given a black and white image of a race track with 16 points marked on it. im supposed to mark these points with a red dot. how can i do this?

thank you in advance

[Geoff Hayes]

Tash - do you know the coordinates of these 16 points, or are you supposed to determine the points using an algorithm? Please clarify your question.

If you know the points/coordinates, then you could use

 plot(X,Y,'ro')

which will plot a red circle given the coordinates defined in the X and Y arrays.

[Tash Dee]

hi Geoff, no im supposed to find the coordinates using an algorithm.

Answer 1

[Image Analyst]

Do you want to add color to an area (like your subject says), or to points (like the body of your question says)? If it's to points, just use

plot(x,y,'r.', 'MarkerSize', 50);

Or if it's to areas, see my attached demos.

[Tash Dee]

hi no i want to colour the points. how am i supposed to locate them using an algorithm?

[Image Analyst]

Then you want to use plot(). But if you're missing the image also, like we are, then it's going to be extraordinarily difficult, if not impossible , to locate points at the edges of a race track. It's tough to do image processing while "blind". All I can suggest is that you threshold and then use bwboundaries(), then try to find 16 points at key locations.

[Tash Dee]

This is the image, i want to colour the numbered points.

[Tash Dee]

[Image Analyst]

It's not too hard. See my Image Segmentation Tutorial http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862. Basically these are the steps:

1. threshold the image: binaryImage = grayImage > 128
2. label the image with bwlabel
3. measure the blobs with regionprops - get area, perimeter, and centroid
4. size filter the blobs to get blobs of only the size of the small white spots.
5. Get rid of the spots in the numbers 6, 8, 9, and 10 by finding the larger round blobs and filling them in.
6. If you want the red spots in the overlay, use plot.
7. If you want an RGB image, then create separate R, G, and B color channels, and then use cat() to create an RGB image of just the spot