Specifications

Image Architecture
2-5
2
applicable ROIs. For example, if an image has both types of ROI and a
COI, operations are performed only on the values of this COI, and only for
those pixels that belong to the intersection of mask ROI and rectangular
ROI.
Both the source and destination image can have a region of interest. In such
cases, operations will be performed on the intersection of the ROIs. Thus,
an image region of interest specifies some part of an image or the entire
image. Once set, the region information of the image remains the same
until changed by the function
SetROI().
NOTE. Not all functions support mask ROI. For example, FFT functions
use only rectangular ROI and COI even if you specify a mask ROI.
Setting an ROI for Multi-Image Operations
Figure 2-1 illustrates image processing operations that take one or more
input images and store the results onto an output image. (Mask ROIs are
not set for the images in this figure.) Before performing any operations,
each function checks that the ROI sizes and offsets are positive. However,
not all functions check that the ROI is within the actual image borders.
All images (input and output) in Figure 2-1 have rectangular ROIs that
specify either the entire image or specific regions set by the
SetROI()
function. The first step is to align the rectangular ROIs of all the images so
that their top left corners coincide. The operation is, then, performed in the
rectangular region where all the images overlap. This scheme gives much
flexibility, effectively enabling translation of image data (even for equal-
size images) from one region of an input image to another region of an
output image.
To successfully perform an image processing operation, one of the
following conditions must be met for the channel of interest (COI):
• Each image (input and output) has one COI,
• Each image (input and output) has all channels included in the ROI
(COI = 0) and all images (input and output) have the same number of
channels (one or more).