Optical-flow imaging for determination of gravy viscosity and uniformity
Optical-flow imaging technology using AI-assisted analysis of multi-angle, multiple-camera images to measure fluid viscosity and consistency in near real time. Non-intrusive and deployable in industrial flow situations, it detects deviations in viscosity and uniformity by analyzing three-dimensional surface flow patterns, enabling inline quality control without sampling.
Overview
This technology leverages optical-flow imaging, a class of imaging methods that measures flow patterns to infer fluid properties such as viscosity and consistency. At low speeds and high viscosities, fluids follow predictable patterns determined by their rheological properties. By applying AI-assisted image analysis to multi-angle, multiple-camera images, the system backtracks these fluid properties from the observed three-dimensional (horizontal plus shallow-depth) surface flow pattern.
The method is non-intrusive and operates in near real time: a connected processing laptop outputs results based on camera feeds positioned at offset angles. Because gravy and similar fluids pass through fixed flow processes during production and application, any deviation in viscosity or consistency appears as a change in the three-dimensional flow pattern, enabling continuous inline quality monitoring.
Technical specifications
The system builds on well-established optical-flow analysis algorithms and extends them with AI-based image analysis for industrial-scale application. Key features:
- Multi-angle cameras capture the surface flow pattern from multiple perspectives to reconstruct horizontal and shallow-depth motion
- AI-assisted processing backtracks fluid viscosity and consistency from the observed flow pattern
- Near real-time results output on a connected processing laptop
- Non-intrusive measurement with no physical contact with the fluid
- Applicable to dynamic flow situations; extendable to static layers by comparing the final layer shape with the expected fluid distribution
Technology readiness level
The technology is at an early validation stage. A three-month validation plan covers: setting up a simple, quick-setup geometry to test the method with actual gravy or a surrogate fluid; comparing results against standard viscosity measurement methods; and introducing artificial non-uniformities to verify detection capabilities. The plan concludes with a report on test results and implementation pathways to industrial-scale processes.
About Arizona State University
Arizona State University is a comprehensive public research university with a multi-campus presence across the Phoenix metropolitan area and a scale that supports interdisciplinary, use-inspired discovery. Industry partners access co-located laboratories, a research and technology park, and innovation centers that house corporate teams with faculty to speed prototyping and validation. A formal alliance with a major hospital system and proximity to a fast-growing manufacturing corridor enable clinical translation and pilot-scale testbeds, while applied student engagements create dependable talent pipelines. Research is backed by competitive federal funding from agencies such as NSF, NIH, DOE, DOD, and NASA. A dedicated technology transfer office supports IP, licensing, and startup formation.
delivered to your inbox.