Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Quality and Safety Assessment Research Unit » Research » Publications at this Location » Publication #375514
Research Project: Develop Rapid Optical Detection Methods for Food Hazards

Location: Quality and Safety Assessment Research Unit

Title: Hyperspectral microscope imaging for food safety and quality applications

Author
item Park, Bosoon
item CHO, JEONG-SEOK - Forest Service (FS)
item KANG, RUI - Nanjing Agricultural University
item EADY, MATTHEW - Fhi360
item OUYANG, QIN - Jiangsu University
item Yoon, Seung-Chul
item Gamble, Gary

Submitted to: ASABE Annual International Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 3/20/2020
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Since a hyperspectral imaging (HSI) technology has been introduced in agriculture and food, many researches for food quality and safety assessment were accomplished successfully, mostly with a prism-grating-prism line scan platform. Furthermore, HSI technology has been employed in micro-scale image acquisition called hyperspectral microscope imaging (HMI) for foodborne pathogen detection specifically live bacterial detection at the cellular level without labels. Thus, we have taken advantage of imaging spectroscopy for evaluation of foods with micro-scale as well as macro-scale image acquisition. In this presentation, two different HMI platforms, acousto-optic tunable filter (AOTF) and Fabry-Perot interferometer (FPI), are employed for food quality and safety evaluation. Performance of those two HMI platforms is compared for foodborne pathogen detection. Also, efficient sample preparation methods are presented for acquiring high quality HMI hypercubes. Additionally, modeling strategy with artificial intelligence (AI) algorithms including convolutional neural networks (CNN) is discussed for classification of agricultural commodity for safety and quality evaluation. Finally, several case studies including foodborne pathogenic bacterial classification, chicken breast meat myopathy detection, green tea quality evaluation, and blueberry firmness measurement at microscopic and cellular level using AOTF-HMI and FPI-HMI are demonstrated. This presentation will expand current hyperspectral imaging technology with integration of microscopy and deep learning algorithms to evaluate food and agricultural commodity for better understanding at cellular level to maintain quality and safety. Keywords: Imaging spectroscopy, Acousto-optic tunable filter, Fabry-Perot interferometer, Food pathogen, Artificial intelligence, Deep learning