Novel Non-Invasive Meat Quality Determination from Machine Learning

Back to all technologies
Download as PDF
2021-KIM-69254
Researchers at Purdue University have developed a new approach to consistently determine meat quality. Currently, a challenge remains in assuring meat quality because traditional equipment for analyzing meat samples is often bulky, expensive, and invasive. Purdue researchers introduce a novel audio sound propagation system that uses mechanical acoustics and machine learning to determine the extent of the microstructural changes in fresh meat products. The new technique allows for noninvasive rapid on-site determination of meat tenderness. By validating the system with pork and beef loin samples, a neural network was established to detect meat quality with minimal latency and improved accuracy. This low-cost method can provide peace of mind in the meat packing process for wholesale packers, retailers, small butchers, hotels, restaurants, and consumers.

Technology Validation: A convolutional neural network (CNN) has been developed for pork and beef loin samples.

Advantages:
-Noninvasive
-Low-Cost and Easy to implement
-Accurate On-Site Determination of Meat Quality

Potential Applications:
-Meat Processing
-Food Science
Oct 23, 2020
Provisional-Patent
United States
(None)
(None)
Purdue Office of Technology Commercialization
The Convergence Center
101 Foundry Drive, Suite 2500
West Lafayette, IN 47906

Phone: (765) 588-3475
Fax: (765) 463-3486
Email: otcip@prf.org