How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
As color machine vision slowly, but steadily, evolves (see page 10), software naturally begins to play a more integral role. Today’s computers are processing color’s much larger data signals increasingly faster. Many software manufacturers are developing new algorithms and tailoring existing libraries to more efficiently meet the need.
Stemmer Imaging’s Common Vision Blox software, for example, is being used in a variety of applications. One example: inspecting oranges in Spain. To meet the need, a number of issues had to be solved.
To meet the need, Stemmer’s Common Vision Blox Image Manager was used with two JAI cameras (M77 color and A11 monochromatic), a Dalsa PC2-Vision frame grabber and StockerYale illumination to identify the color, size and shape of the oranges.
In the system, two cameras are placed in two different positions on the line. Angled mirrors are placed in the field of both cameras to image three sides of each orange. The first camera is an RGB color model that uses an external trigger to capture the image. This camera determines the color of each piece of fruit, with four oranges captured on each image. The second camera is a monochrome model that also acquires four oranges per image and is used to calibrate the fruit.
“With the growing availability of software tools for both the analysis and processing of color images, as well as affordable yet high-quality color cameras, the color information becomes more and more noticed by the user,” says Volker Gimple, Group Manager, Software Development, at Stemmer Imaging. “Color now is an integral part of the image information. As a matter of fact, state-of-the-art algorithms for classification and object recognition like Common Vision Blox, utilize color as a decision criterion equivalent to geometric or brightness information.”