Evaluating geometric pattern matching

The first step in any machine vision application -- and the one that usually determines whether the application succeeds or fails -- involves locating the object within the vision camera's field of view, a process known as pattern matching. Pattern matching can be extremely challenging, as many variables can alter the way an object appears to a vision system. Traditional pattern matching technology relies upon a pixel-grid analysis process commonly known as normalized correlation. Cognex Corp. developed a new technology called geometric pattern matching, which learns an object's geometry. As other vision companies have interpreted the technology, specific products vary widely. This white paper from Cognex can guide you through ten of the most important questions regarding pattern matching yield, accuracy, speed and pattern training.

Cognex's geometric pattern matching, marketed under the brand name of PatMax, learns an object's geometry using a set of boundary curves that are not tied to a pixel grid, and then looks for similar shapes in the image without relying on specific gray levels. The result is a significant improvement in the ability to accurately find objects despite changes in angle, size and shading.

• Email This Page
• Print This Page