What would the world look like in 2D? If objects had height and width but no depth, everything would appear to you as flat shapes, with no thickness. Picking up your cup of coffee in the morning and bringing it to your lips would be a much more time-consuming affair.
You’d have to actively move your head around the cup and really think about stitching together the different dimensions of it in your mind, before picking it up and taking your first sip.
Simpler and easier to see things in 3D, right?
Now, imagine trying to do things as complicated as automotive assembly and materials handling with just two dimensions. With tasks that complex, the limitations are powerful.
This is why Liberty Robotics uses At-a-Glance, full field, 3D geometry. Just as your eyes can take in your entire coffee cup at once, our vision systems see the entire scene in front of them at once. That gives us 3 key advantages:
1. Speed: This one is the most powerful. Because we acquire the full field of geometry at once, we simultaneously acquire all of the recognizable objects in the scene at once. We know the dimensions immediately, without needing time to decipher the scene in pieces. Our 2D competitors have to stitch the scene together, and this adds cycle time.
2. Accuracy: 2D systems are ideal as long as you never move objects closer or further from the sensor and nothing ever tilts. That’s just not the world we live in. The reason for this is that 2D systems measure distances in pixels, but when objects do in fact move closer or father away from the sensor, the distances can no longer be measured in whole pixels.
This best way to explain this is by example. Let’s say you have a 2D sensor measuring a 50 millimeter object that is one meter away. Here, 1 pixel equals 1 millimeter. If that object moves up or down, left or right, by one millimeter, the sensor sees the image of the cutout change by 1 full pixel. That is a detectable change. However, if the cutout moves closer or further from the sensor by one millimeter, the image of the cutout changes by just a fraction of a pixel. Such a small change is very likely undetectable.
Or you can just measure in combined 2D + 3D like Liberty Robotics does.
3. Flexibility: This is where the biggest cost savings advantage of 3D comes in. Our systems can adapt to model changeovers, the addition of new models, and novel objects the system hasn’t seen before without having to reinstall or upgrade the system. We can even measure parts in motion.
Liberty Robotics 3D vision can be adapted easily to accommodate a broader variety of part and angle presentations than exclusively 2D vision approaches. Once it’s installed, there is no need for additional hardware to accommodate new model styles, and retraining for new items can be done in as little as a few hours. All of this translates into cost savings.
Using the patented Information Theoretic approach that Liberty Robotics has pioneered, which you can see here, we can combine both the pixel measurements of 2D with 3D depth measurements to create a complete picture in less time, with greater accuracy, and without new hardware additions in the future.
If you’re ready to discuss what Liberty Robotics can do for you, click here to get the conversation started. Follow us on LinkedIn for the latest news.