Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Environmental Microbial & Food Safety Laboratory » Research » Research Project #440122

Research Project: Embedded AI Techniques for Compact Optical Sensing Systems for Food Safety

Location: Environmental Microbial & Food Safety Laboratory

Project Number: 8042-42000-021-005-S
Project Type: Non-Assistance Cooperative Agreement

Start Date: May 1, 2021
End Date: Apr 30, 2026

Various multimodal spectral imaging technologies have been developed to rapidly and nondestructively assess safety attributes of food products. Spectral imaging creates a large volume of data that requires complex processing and analyses. The objective is to develop artificial intelligence (AI) techniques for spectral and image processing, analysis, and modeling, which can be embedded in ARS next-generation portable and compact sensing devices and systems for automated real-time target detection and intelligent food safety assessment useful to non-expert users.

The sensing technology team of EMFSL is develop handheld hyperspectral fluorescence imaging devices for food contamination and sanitation inspection in food processing environments, and a compact automated hyperspectral imaging system. The handheld device will be used to collect hyperspectral fluorescence images from a broad range of organic residues and food contaminants commonly found during sanitation inspection and audit in food processing facilities. The hyperspectral image data obtained from the imaging systems will be shared with Cooperator to develop spectral and image processing algorithms and machine learning and deep learning classification models (e.g., support vector machines and convolutional neural networks) using commercial and/or open-source artificial intelligence (AI) software libraries (e.g., MATALB, TensorFlow and PyTorch). The developed algorithms and AI models will be embedded in ARS in-house developed software of the imaging systems, which will be used for real-time detection and visualization for food contamination and sanitation inspection.