[mmdpotatoes] [Up] [mmdruler] Demonstrations

mmdrobotop
Detect marks on a robot.

Description

The input-image is a gray-scale image of a rounded-shaped robot, viewed by a camera mounted on the ceiling. The procedure detects two white marks on top of the robot. This is a typical example of application of the top-hat operator in image segmentation.

Demo Script

Reading

The gray-scale image of the robot top view is read.

>>> a = mmreadgray('robotop.tif');

                  
>>> mmshow(a);

                
a

Open top-hat

It detects white regions smaller than a square of radius 4.

>>> b = mmopenth(a,mmsebox(4));

                  
>>> mmshow(b);

                
b

Opening

It removes white objects smaller than a square of radius 1.

>>> c = mmopen(b,mmsebox());

                  
>>> mmshow(c);

                
c

Thresholding

It detects the robot markers. This is a very robust thresholding (i.e., the result is not sensible to small changes in the value of the threshold parameter). The original image is overlayed by the detected robot markers.

>>> d = mmthreshad(c,100);
Warning: Converting input image from int32 to uint8.
>>> mmshow(a,d);

                
a,d

[mmdpotatoes] [Up] [mmdruler] Python