MOSA COMMAND DEMONSTRATION

The MOSA command is used primarily to put together several shots of the same object in order to have a single image of this object. To illustrate the use of this command, we will create a mosaic from two wide field images of the moon: PLATO.FIT and COPERNIC.FIT (download and decompress the file MOSA.ZIP) . The first one covers the northern part of the moon; the other, the central part.

Let's look at these two images:

LOAD PLATO
VISU 3000 700
LOAD COPERNIC
VISU 3000 700
 

Plato region (left) and Copernic region (right).

The first thing to do is to find a well contrasted landmark that is common to both images.  Here, this is the crater PYTHEAS, which is near the coordinates (274,151) on the image PLATO.FIT and near (4,138) on COPERNIC.FIT.  These coordinates were found with the cursor.

The shift between the two images is:

· along X: 274 - 4 = 270 pixels
· along Y: 151 - 138 = 13 pixels

(note that the relative position between images can be a fractional number).

Now type:

LOAD PLATO
MOSA COPERNIC 270 13 0

Result of the command MOSA 270 13 0.

The overlapping zone is noticeable because the intensity level of the two images is not identical. To improve the result, reexecute the MOSA command, by setting the last parameter (overlap type) to 5:

LOAD PLATO
MOSA COPERNIC 270 13 5

Result of the command MOSA 270 13 5.

 
The value of the overlap type allows you to choose the state of the overlapping zone in the final image:

• If overlap type = 0, the second input image (COPERNIC in the example) overwrites the first input image (PLATO in the example) unless the pixels in the second input image have the intensity zero.  In this case, the values from the first input image are used.
• If overlap type = 1: the first input image overwrites the second input image unless the pixels in the first input image have zero intensity, in which case the values from the second input image are used.
• If overlap type = 2: the pixels in the output image take the maximum value of the pixels from the two input images.
• If overlap type = 3: the pixels in the output image take the minimum value of the pixels from the two input images.
• If overlap type = 4: the pixels in the output image take the average value of the pixels from the two input images.
• If overlap type = 5: a bilinear interpolation is performed on the common parts of the two input images.

Finally, it is of course possible to link several MOSA commands together to create very large mosaics from more than two images.

You can also use a dialog box for mosaic operations. For more details click here.