pixelate / pixel search

[systems_eng]

Ultimately, I'm trying  to analyze a topographic map, to locate elevated areas which are defined by light spots on a 256 grey scale, then apply a coordiate (x/y) to that spot.  I've read some of the posts on similar topics but, I am still unsure if I'm using the thought process to solve the issue.

I start by receiving a 256 grey scale map.  My thoughts are to...
pixelate the image
isolate the high peaks (light color)
smooth out highlighted spots (close gaps / remove noise)
locate the pixels of specified color
apply a position to pixels

First off, I'm not sure if some of these steps are neccessary.  Second, I'm not entirely sure how to locate the pixel and specify a coordinate.  I've seen tutorials on applying a distance to the a specific spot, but not one that directly provides a solution to a problem such as this.

[systems_eng]

Looks like I'm replying to my own post again...

The way I went about solving this issue was to put a 256 grey-scale topography graph through several filters in Vision Assistant:
Brightness - Color Isolation
Extract HSL - Luminance Plane
Smoothing - Gaussian
Close - Removes holes and noise
Brightness - Binary Threshold

From here I used the Detect Objects function to locate all my spots of interest (those that made it through all the filters), then displayed the calculated (x,y) coordinates with the Calculator function.  However, I am still stuck with the problem of displaying the coordinates of multiple points.  As of now, the program will only display the coordinates of the first object that it encounters.

[David S.]

Hi systems_eng -

You may be able to come up with a more efficient algorithm by thresholding earlier in the process and cleaning up the binary image instead of the grayscale one.  If you can post a sample image, I'd be interested in tinkering with it myself.

Which function are you talking about with "Detect Objects"?  This isn't a step in Vision Assistant 8.0 or Vision Development Module 8.0.

You can display your resultant coordinates  on the image itself with the Overlay functions once you export the script to a LabVIEW program.  IMAQ Overlay Text should be most useful here.