Rockwell Automation 5370-CVIM2 Module - Manual (Page 296)

Chapter 7
Inspection Tools
7–58
Figure 7.49 Example: Image Tool and Window in Toolset Edit Panel
An image tool can be configured for one of three principal types of image
enhancement operations; namely, transform, convolve, and image arithmetic
operations.
Note that morphology filtering is available to all three image enhancement
operations (see Chapter 8, Threshold, Filters, And Morphology for details
about morphology filtering). Special look up tables (LUTs) are also available
to the three image enhancement operations; however, they are not normally
used with the transform operation.
Here is a brief description of each image enhancement operation:
Transform –– A transform operation can perform spatial transformation
(“warping”) functions, which alter the shape of the image within an image
tool, but do not alter the gray scale values of the pixels in the image. These
functions are available only with the arc ring, quad, and perspective shape
selections; they are not available with the convolve and image arithmetic
operations. (The Shape section on page 7–87 of this manual provides details
about spatial transformation operations.)
Convolve –– A convolve operation can perform one of several spatial
filtering functions, each of which provides one of the following basic results:
•
Enhances gradients in the image that represent edges (or flaws) on
inspected objects.
•
Removes noise from the image.
Spatial filtering functions employ a “kernel” that performs “neighborhood
operations” on all pixels within an image tool’s “area of interest.” The kernel
is a matrix of coefficients that perform calculations on the gray scale values
of a “neighborhood” of pixels, and the result of these calculations modifies
the pixel that lies under the center of the matrix. This process continues
throughout the image, moving from left to right and top to bottom.
Overview: Image Tool
Operations