AI Pioneer Andrew Ng Headlines Vision Spectra Conference ’22

By harnessing the power of “good” data over “big” data, manufacturers with limited data sets can use machine vision for quality inspection. This is the premise of the keynote address to be delivered by globally recognized AI visionary Andrew Ng for the second annual Vision Spectra Conference (VSC), which will be held online July 19-21.

The founder and CEO of Landing AI underscores how most AI models rely on huge data sets such as those amassed by Google, Facebook, and Amazon. But manufacturers do not always have access to the large amount of data that is needed to feed AI engines to target specific issues. By using the data-centric AI approach that Ng pioneered, 50 good images can be as effective as 500 bad ones at detecting manufacturing defects.

Ng’s address kicks off the three-day conference, which brings together technology innovators in inspection quality control, deep learning, logistics, and other prominent industry sectors. Across more than 30 sessions, the conference program explores advancements in vision-guided robotics, cameras and sensors, and industrial inspection.

In a second keynote, Luca Verre, co-founder and CEO of Prophesee, presents on neuromorphic sensing. Neuromorphic sensor pixels only report when they sense movement, meaning that the independence of each pixel enables the sensors to overlook irrelevant information and thereby produce up to 1000× less data than a conventional sensor, while achieving a higher equivalent temporal resolution of greater than 10,000 fps. Applications include scientific analysis, robotics, and industrial automation.

In a third keynote, Richard Liu of Yole Développement offers a macro look at the machine vision market and identifies market demands and drivers, as well as long- and short-term growth factors. Imaging trends, dictated by camera technologies and modules, will be a focus of the session. He will also cover global demand, emerging markets, and various imaging modalities.

The conference is offered exclusively online. Registration and attendance are free.

Five program tracks

First-time presenters will join companies and session leaders who are returning to the VSC. Presentations are split across five program tracks. Photoneo, SICK AG, FANUC, and LUCID Vision Labs are among the companies contributing to the Vision-Guided Robotics track.

In the Inspection and AI track, Neurala presents on the topic of vision for AI industrial inspections, and Hamamatsu Corp. discusses food and beverage inspection. In the Cameras and Sensors track, Princeton Infrared Technologies covers SWIR sensing, and imec presents on quantum dot sensing.

The following topics and more will be featured in sessions in a track exploring Warehouse and Logistics and the supply chain: CapSen Robotics and Artemis Vision examine vision-guided innovations for warehouse environments. There will also be a Components track.

Returning presenters include David L. Dechow, who was recently named vice president of outreach and vision technology at Landing AI; Nick Sischka and Greg Hollows from Edmund Optics; Tom Brennan from Artemis Vision; and Jonathan Hou from Pleora.

Hou will target manufacturers who rely on human inspection at key production stages. In his presentation, he discusses how the development of new “no code” algorithms and training tools, when paired with turnkey inspection systems, is increasing access to AI in all areas of manufacturing.

Register now

The Vision Spectra Conference is the third conference in Photonics Media’s 2022 lineup. Registration and additional information are available at www.photonics.com/vsc2022.