Presentation: 2025 ND EPSCoR Annual conference 

October 21, 2025, NDSU Memorial Union, Fargo, North Dakota

Hyperspectral Imaging and Machine Learning Applications for Specialty Crops

Lettuce is one of the most widely grown vegetables and serves as an important source of nutrients in the human diet. Therefore, estimating nutrient components in lettuce during the early growth stage is crucial. Traditional methods, such as laboratory tests, are often expensive and time-consuming. Hyperspectral imaging (HSI) provides a cheaper, faster, and non-destructive alternative to these traditional methods. When combined with artificial neural networks (ANN) and various machine learning methods, hyperspectral imaging can offer effective, accurate, and robust estimation of nutrient components in lettuce. On the other hand, HSI data are often noisy and large in volume due to the ineffective analysis of lettuce reflectance. Therefore, it is important that the most suitable region of lettuce leaves is identified to obtain most appropriate data from HSI. In this study, reflectance values from different regions of lettuce leaves will be analyzed and compared to determine the most suitable region that provides the most accurate and robust estimation of nutrient components.