Stevens Institute of Technology helps AI reach out and touch someone

19 Nov 2024

Lidar-based method uses back-scatter from surfaces to measure texture and roughness.

AI-based technologies are increasingly able to gather data from the world around them as the basis for their calculations, by interfacing with relevant sensing platforms.

These sensors allow AIs to observe the environment, to converse and to calculate. But so far one thing they have not done well is "feel" surfaces - measure texture and roughness, a purely mechanical function.

A project at Stevens Institute of Technology has now demonstrated a novel laser-based approach to surface roughness metrology, which might also be suitable for transferring that data into an AI system for evaluation.

Published in Applied Optics, the research may lead to valuable advances in metrology for medicine, manufacturing and other applications.

"AI has more or less acquired the sense of sight, through advances in computer vision and object recognition," said Stevens Institute physics professor Yong Meng Sua. "It has not, however, yet developed a human-like sense of touch that can discern, for example, a rough sheet of newspaper paper from a smooth and glossy sheet of magazine paper."

The researchers devised a quantum-optics setup that combines a photon-firing scanning laser with new algorithmic AI models trained to tell the differences among various surfaces as they are imaged with those lasers - a "marriage of AI and quantum," commented Daniel Tafone from Stevens.

The platform uses a single-pixel raster scanning and single-photon counting lidar system directed at a surface in picosecond pulses. Reflected back-scattered photons return from the target object carrying speckle noise, normally considered detrimental.

However, the Stevens group’s system detects and processes the noise artifacts using an AI that has been trained to interpret their characteristics as valuable data. This allows the system to accurately discern the topography of the object.

Spotting skin cancers early

"These back-scattered photons carry speckle noise produced by the rough surface, and the variation in photon counts over different illumination points across the surface becomes a good measure of its roughness," commented the project in its published paper.

In trials using 31 industrial sandpapers with surfaces of varying roughness, the group’s method averaged a root-mean-square error of about 8 microns, according to the team. After working with multiple samples and averaging results across them, its accuracy improved significantly to within 4 microns, comparable to the best industrial profilometer devices currently used.

"Interestingly, our system worked best for the finest-grained surfaces, such as diamond lapping film and aluminum oxide," commented Tafone.

Since lidar technology is already widely implemented across autonomous cars, smartphones and robots, the new metrology approach could enrich the capabilities of those platforms to perform surface property measurements at very small scales.

Applications for the new surface metrology system could also go beyond manufacturing. The project commented that diagnosis of skin cancers can be hindered by mistakes made by the human examiners who confuse very similar-looking but harmless conditions with potentially fatal melanomas.

"Tiny differences in mole roughness, too small to see with the human eye but measurable with our proposed quantum system, could differentiate between those conditions," said Yuping Huang from Stevens Center for Quantum Science and Engineering. "Quantum interactions provide a wealth of information, using AI to quickly understand and process it is the next logical step."