The problem with transparency

MICHAEL D. WHEELER, EDITOR-IN-CHIEF, MICHAEL.WHEELER@PHOTONICS.COM

An exponential growth in online retail sales has occurred since the turn of the millennium, as a growing portion of the world’s population has gravitated toward e-commerce. This growth has spurred a rise in robotics and automation, with the integration of robotics in fulfillment centers playing a fundamental role in the continual improvement of their capabilities, efficiency, and consistency.

If you think about your recent online orders, it is likely that they arrived enclosed in a poly bag, bubble packaging, or shrink-wrap. 3D-imaging such transparent or semi-transparent items has long been considered the ultimate challenge. The difficulty lies in their unique optical properties, complicated by the fact that light either passes right through them or reflects off them in uncontrollable directions. Also, consider that if an object is viewed in a poly bag, multiple surfaces will be seen — the surface of the bag, the surface of the object enclosed, and the surface of the back of the bag.

Solving this challenge is the subject of two separate pieces in this edition. Zivid’s Martin Ingvaldsen, John Leonard, and Henrik Schumann-Olsen share how they used a sophisticated 3D vision system and structured light to generate detailed point clouds. These point clouds could be used for efficient object recognition and handling. “3D Vision Transforms Robotic Piece Picking by Imaging Transparent Materials” starts here.

And in “Transparent, Shiny Materials Captured with 3D Scanning,” contributing writer Dominic Acquista examines how Photoneo tackled the same problem. In the past, companies would scan transparent objects by adjusting their scanner’s position or even take the extreme step of covering objects in special coatings and sprays to mitigate unwanted effects of light. But such methods proved to be impractical.

Photoneo’s approach involves technology that effectively scans and recognizes reflective and transparent objects using a proprietary CMOS computational image sensor and the company’s parallel structured light technology. Multiple patterns are captured in a single laser sweep, enabling transparent scanning to occur a magnitude faster than using the standard approach. For more details, click here.

Finally, for the very latest in the different modalities of 3D imaging technology, don’t miss David Dechow’s “The Evolution of 3D Imaging in Industrial Applications” here. Learn about innovations in techniques, the evolution in lighting, and emerging applications.

I hope you enjoy the issue!
>