R. Rao, M. Salas, S. Singh, A. Mittal, J. Yarbrough, A. Singh, L. Pal
North Carolina State University,
United States
Keywords: food waste, hyperspectral imaging, landfill, contamination, sustainable development goals
Summary:
In the United States, 30–40% of the food supply is never consumed, representing tens of billions of pounds of wasted food each year. This growing food waste stream poses environmental and economic challenges, including landfill burden and greenhouse gas emissions. This study presents a hyperspectral imaging (HSI)-driven framework for rapid, non-destructive characterization of heterogeneous food waste. HSI was used to capture high-resolution spectral information across diverse waste fractions such as fruits, vegetables, meats, and starch-based materials to distinguish clean from contaminated samples and measure compositional variability. Spectral features were analyzed to assess contamination, moisture variation, and chemical heterogeneity, all of which are critical for maintaining consistent feedstock quality for downstream valorization. Each food waste category exhibited unique spectral fingerprints, showing its inherent composition and surface condition. By linking spectral data with selective laboratory analyses, the framework establishes a robust foundation for automated waste classification and targeted conversion into bioenergy, compost, and biochemicals such as bioethanol, lactic acid, and bioplastics. Our HSI-based characterization approach supports landfill diversion, improves waste valorization efficiency, and advances the United Nations Sustainable Development Goals related to responsible consumption and climate action.