Collaboration brings machine learning to multispectral and hyperspectral cameras
PR Sys Design and Cubert have announced a partnership that brings machine learning technology to multispectral and hyperspectral cameras.
Developed by PR Sys Design, perClass is a software tool that enables research and development specialists to create classification solutions. With a Matlab-based toolbox within perClass, users can identify a problem, interactively work with data, choose features, train classifiers, and optimize performance. Further, with the perClass software development kit, users can utilize perClass Runtime to embed classifiers in applications running without Matlab.
With the collaboration, users can create classification solutions for Cubert’s real-time spectral cameras. Within minutes, according to a press release from PR Sys Design, users will be able to record spectral images, create statistical classifiers for specific materials, and deploy such solution processing a live data stream. This, according to PR Sys Design, enables users to identify things such as types of plastics, sort natural products such as vegetables or nuts, classify types of tissues and biological material, analyze tree types and plant health in remote sensing images, and more.
perClass software comes integrated within the 2.0.4 “Fuchsia” release of Cubert utilities. Users can install the perClass Toolbox for Matlab directly with the Cubert installer and multispectral and hyperspectral images acquired by Cubert cameras can be directly imported into the perClass environment. Additionally, a set of interactive visualization data modeling and classification tools are available for rapid prototyping.
Learn more: search the Vision Systems Design Buyer's Guide for companies, new products, press releases, and videos